An adaptive polarization navigation decision method and related device

Abstract

The application discloses a kind of adaptive polarization navigation decision-making method and related equipment, method includes: obtaining the original light field environment parameter of underwater equipment, original light field environment parameter is preprocessed, obtain target light field environment parameter set;Light field scene recognition is carried out based on target light field environment parameter set, obtain current light field scene type;According to current light field scene type, the dynamic change of light field environment parameter is combined, the benchmark decision parameter corresponding to current light field scene type is dynamically iterated and optimized, and optimal decision parameter is obtained;According to current light field scene type, optimal solution algorithm is determined, and the polarization light parameter in target light field environment parameter set is solved in combination with optimal decision parameter, and navigation data is obtained;Navigation data is checked, and navigation decision instruction that passes through checking is output.The application can greatly improve the accuracy and stability of navigation decision in complex underwater light field environment, and can be widely applied in underwater autonomous navigation technical field.

Description

Technical Field

[0001] This application relates to the field of underwater autonomous navigation technology, and in particular to an adaptive polarization navigation decision-making method and related equipment. Background Technology

[0002] Currently, underwater navigation technology has become a core support for fields such as marine resource exploration, national defense security, and deep-sea scientific research. Among them, polarized light navigation, with its unique advantages of being passive, having no cumulative error, resisting electromagnetic interference, and operating in all weather conditions, is gradually replacing traditional inertial navigation and underwater acoustic navigation, becoming one of the mainstream technologies for high-end underwater navigation. The core principle of polarized light navigation is to utilize the propagation characteristics of solar polarized light in the atmosphere and water bodies. By sensing parameters such as the vibration direction and degree of polarization of the polarized light, the heading angle and attitude angle of the carrier are calculated to achieve high-precision positioning and navigation. However, the underwater optical field environment has extremely high complexity and uncertainty, which seriously restricts the accuracy and stability of polarized navigation.

[0003] Most polarization navigation software decision-making methods and systems are based on fixed decision parameters and a single solution algorithm, which makes it difficult to adapt to the changing underwater optical field, easily resulting in large solution errors and insufficient navigation accuracy and system reliability. In addition, the processing links of related systems are separated, data transmission is delayed, and decision response is lagging.

[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0005] The embodiments of this application aim to at least partially address one of the technical problems in the related art. Therefore, the main objective of the embodiments of this application is to propose an adaptive polarization navigation decision-making method and related equipment, which can significantly improve the accuracy and stability of navigation decisions in complex underwater optical field environments.

[0006] To achieve the above objectives, one aspect of this application proposes an adaptive polarization navigation decision-making method, the method comprising the following steps: The original optical field environment parameters of the underwater environment in which the target underwater device is located at the current moment are obtained, and the original optical field environment parameters are preprocessed to obtain the target optical field environment parameter set. Based on the target light field environment parameter set, light field scene recognition is performed to obtain the current light field scene type; Based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, the baseline decision parameters corresponding to the current light field scene type are dynamically iteratively optimized to obtain the optimal decision parameters. The optimal solution algorithm is determined based on the current light field scene type, and the polarization parameters in the target light field environment parameter set are solved in combination with the optimal decision parameters to obtain navigation data. The navigation data is verified, and a navigation decision command that passes the verification is output.

[0007] In some embodiments, the preprocessing of the original optical field environment parameters to obtain the target optical field environment parameter set includes: The original optical field environment parameters are subjected to outlier removal processing to obtain the first optical field environment parameters; The first optical field environmental parameters are subjected to noise filtering to obtain the second optical field environmental parameters; The second light field environment parameters are subjected to data standardization processing to obtain the target light field environment parameter set.

[0008] In some embodiments, the step of identifying the light field scene based on the target light field environment parameter set to obtain the current light field scene type includes: The target light field environment parameter set is subjected to feature extraction processing to obtain the current light field environment features; wherein, the current light field environment features include the current polarization degree variance, the current polarization angle fluctuation amplitude, the current light intensity change rate, and the current turbidity value; The cosine similarity algorithm is used to calculate the global similarity between the current light field environment features and the standard scene features in the preset standard scene feature library; If the global similarity is greater than or equal to the preset similarity threshold of the target standard scene, then the target standard scene is determined as the current light field scene type.

[0009] In some embodiments, the step of dynamically iteratively optimizing the baseline decision parameters corresponding to the current light field scene type based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set to obtain the optimal decision parameters includes: Based on the current light field scene type, determine the benchmark decision parameters corresponding to the current light field scene type; The particle swarm optimization algorithm is used to dynamically iterate and optimize the baseline decision parameters by combining the dynamic changes of the light field environment parameters in the target light field environment parameter set, so as to obtain the optimal decision parameters that are suitable for the current light field scene type.

[0010] In some embodiments, the step of verifying the navigation data and outputting a navigation decision instruction that passes the verification includes: The navigation data is fused and verified with the inertial navigation data and underwater acoustic navigation auxiliary data corresponding to the target underwater device to obtain the current polarization navigation error; If the current polarization navigation error is less than or equal to a preset navigation accuracy threshold, then the navigation data verification is determined to be successful, and the navigation decision instruction is generated based on the verified navigation data. If the current polarization navigation error is greater than the preset navigation accuracy threshold, the navigation data is discarded, and the process returns to the step of determining the optimal solution algorithm based on the current light field scene type, and combining the optimal decision parameters to solve the polarization parameters in the target light field environment parameter set to obtain navigation data, until the navigation decision command that has passed verification is output.

[0011] In some embodiments, the method further includes: During the process of light field scene recognition, the light field environment parameters continuously collected from the target light field environment parameter set are linearly fitted to obtain parameter change trend data. Based on the parameter change trend data and the preset parameter change threshold, light field anomalies are identified. If it is determined that there is an anomaly in the light field environment parameters in the target light field environment parameter set, a light field anomaly warning signal for the current light field scene type is output. After receiving the light field anomaly warning signal, the baseline decision parameters corresponding to the current light field scene type are automatically adjusted according to the preset anomaly adjustment strategy; Based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, the adjusted baseline decision parameters are dynamically iteratively optimized to obtain the optimal decision parameters.

[0012] To achieve the above objectives, another aspect of this application proposes an adaptive polarization navigation decision-making device, which includes the following modules: The data acquisition and preprocessing module is used to acquire the original light field environment parameters of the underwater environment in which the target underwater device is located at the current moment, and to preprocess the original light field environment parameters to obtain the target light field environment parameter set. The light field scene recognition module is used to perform light field scene recognition based on the target light field environment parameter set to obtain the current light field scene type. The decision parameter optimization module is used to dynamically iteratively optimize the baseline decision parameters corresponding to the current light field scene type based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, so as to obtain the optimal decision parameters. The polarization parameter calculation module is used to determine the optimal calculation algorithm based on the current light field scene type, and combine the optimal decision parameters to calculate the polarization parameters in the target light field environment parameter set to obtain navigation data. The navigation data verification module is used to verify the navigation data and output navigation decision instructions that have passed the verification.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0015] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides an adaptive polarization navigation decision-making method and related equipment. This scheme obtains the original light field environment parameters of the underwater environment in which the target underwater device is located at the current moment, and preprocesses the original light field environment parameters to obtain a target light field environment parameter set; performs light field scene recognition based on the target light field environment parameter set to obtain the current light field scene type; dynamically iterates and optimizes the benchmark decision parameters corresponding to the current light field scene type according to the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set to obtain the optimal decision parameters; determines the optimal solution algorithm according to the current light field scene type, and calculates the polarization light parameters in the target light field environment parameter set according to the optimal decision parameters to obtain navigation data; performs verification processing on the navigation data and outputs a navigation decision command that has passed the verification. This application's embodiments achieve accurate light field scene recognition based on a target light field environment parameter set, effectively distinguishing different underwater light field conditions and realizing intelligent perception and discrimination of complex underwater light field environments. This provides accurate scene basis for subsequent parameter adaptive optimization and algorithm matching switching. Based on the identified current light field scene type and combined with the real-time dynamic changes of light field environment parameters, the corresponding benchmark decision parameters for the scene are dynamically iteratively optimized. This adapts to the real-time changing characteristics of different underwater light field environments, overcoming the shortcomings of traditional fixed benchmark parameters that cannot adapt to complex dynamic underwater scenes, ensuring optimal adaptability under different conditions. Superior navigation decision parameters significantly improve the adaptability and accuracy of navigation environments. The system matches the optimal solution algorithm to the light field scene type and combines it with optimized decision parameters to complete the polarization parameter calculation. This allows for targeted adaptation to the calculation needs of different light field scenes, significantly improving the calculation accuracy and output stability of underwater polarization navigation. By verifying the calculated navigation data and filtering out compliant and valid navigation decision commands, the system effectively avoids navigation deviations caused by abnormal calculation data output, greatly improving the accuracy and effectiveness of underwater navigation decisions, thereby ensuring the precision and reliability of underwater equipment navigation operations. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of an adaptive polarization navigation decision-making method provided in an embodiment of this application. Figure 2 This is a block diagram of the module structure of an adaptive polarization navigation software decision system for complex underwater optical fields provided in an embodiment of this application; Figure 3 This is a flowchart of an adaptive polarization navigation software decision-making method for complex underwater optical fields provided in an embodiment of this application; Figure 4 This is a schematic diagram of a light field scene recognition process provided in an embodiment of this application; Figure 5This is a schematic diagram of a decision parameter adaptive optimization process provided in an embodiment of this application; Figure 6 This is a logical schematic diagram of a multi-algorithm adaptive switching provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of an adaptive polarization navigation decision device provided in an embodiment of this application; Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application.

[0019] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0020] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] Currently, underwater navigation technology has become a core support for fields such as marine resource exploration, national defense security, and deep-sea scientific research. Among them, polarized light navigation, with its unique advantages of being passive, having no cumulative error, resisting electromagnetic interference, and operating in all weather conditions, is gradually replacing traditional inertial navigation and underwater acoustic navigation, becoming one of the mainstream technologies for high-end underwater navigation. The core principle of polarized light navigation is to utilize the propagation characteristics of solar polarized light in the atmosphere and water bodies. By sensing parameters such as the vibration direction and degree of polarization of the polarized light, the heading angle and attitude angle of the carrier are calculated to achieve high-precision positioning and navigation.

[0023] However, the underwater light field environment is extremely complex and uncertain, which severely restricts the accuracy and stability of polarization navigation. On the one hand, the underwater light field is affected by factors such as water scattering, absorption, and reflection. The polarization parameters (degree of polarization, polarization angle, and light intensity) of different depths and different water areas (shallow / deep sea, turbid / clear water) vary significantly and fluctuate dynamically with time and water flow. On the other hand, there are complex scenarios underwater, such as sudden changes in light intensity (e.g., direct sunlight on surface water, weak light in the deep sea), uneven turbidity (e.g., river estuaries, disturbance of seabed sediments), and interference from external light sources (e.g., lights from underwater equipment, other underwater body light sources). These factors make the polarization light signal prone to distortion and reduce the signal-to-noise ratio.

[0024] Most existing polarization navigation software decision-making methods and systems are based on fixed decision parameters and a single solution algorithm, lacking the ability to adaptively adjust to complex underwater optical fields. Firstly, the decision parameters (such as polarization signal acquisition threshold, number of solution iterations, and filter coefficients) remain fixed, making dynamic optimization based on the real-time optical field environment impossible. This can easily lead to increased solution errors and navigation failures when the optical field changes abruptly or turbidity increases. Secondly, the solution algorithms are often limited to fixed Kalman filtering or least squares algorithms, failing to adapt to different optical field scenarios (such as low-light or high-turbidity scenarios), resulting in unstable navigation accuracy. Thirdly, the lack of real-time perception and prediction capabilities for the optical field environment prevents early identification of optical field anomalies (such as signal distortion or interference intrusion), forcing a passive response and further reducing the reliability of the navigation system. Furthermore, many systems design polarization signal acquisition, solution, and decision-making in a fragmented manner, resulting in high data transmission delays, untimely decision responses, and a lack of fault self-diagnosis and fault tolerance mechanisms. When a module malfunctions, the entire navigation system can easily fail, failing to meet the practical needs of long-term, high-precision navigation in complex underwater scenarios.

[0025] In view of this, this application provides an adaptive polarization navigation decision-making method and related equipment. This scheme completes accurate light field scene identification based on the target light field environment parameter set, effectively distinguishing different underwater light field conditions and realizing intelligent perception and discrimination of complex underwater light field environments. This provides accurate scene basis for subsequent parameter adaptive optimization and algorithm matching switching. Based on the identified current light field scene type and combined with the real-time dynamic changes of light field environment parameters, the corresponding benchmark decision parameters of the scene are dynamically iteratively optimized. This can adapt to the real-time changing characteristics of different underwater light field environments, overcoming the shortcomings of traditional fixed benchmark parameters that cannot adapt to complex dynamic underwater scenes. It ensures that optimal navigation decision parameters are obtained under different operating conditions, significantly improving the adaptability of navigation environments and the accuracy of parameters. It matches the optimal solution algorithm according to the type of light field scene and combines it with the optimized decision parameters to complete the polarization parameter calculation, enabling targeted adaptation to the calculation needs of different light field scenes and significantly improving the calculation accuracy and output stability of underwater polarization navigation. By verifying the calculated navigation data and filtering out compliant and valid navigation decision commands, it effectively avoids navigation deviation problems caused by abnormal calculation data output, significantly improving the accuracy and effectiveness of underwater navigation decisions, thereby ensuring the accuracy and reliability of underwater equipment navigation operations.

[0026] This application provides an adaptive polarization navigation decision-making method, relating to the field of underwater autonomous navigation technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle-mounted terminal, but is not limited thereto. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the adaptive polarization navigation decision-making method, but is not limited to the above forms.

[0027] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0028] Please see Figure 1 , Figure 1 This is an optional flowchart of an adaptive polarization navigation decision-making method provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S105.

[0029] Step S101: Obtain the original light field environment parameters of the underwater environment in which the target underwater device is located at the current moment, and preprocess the original light field environment parameters to obtain the target light field environment parameter set; In some embodiments, the step of preprocessing the original optical field environment parameters to obtain a target optical field environment parameter set may include: performing outlier removal processing on the original optical field environment parameters to obtain a first optical field environment parameter; performing noise filtering processing on the first optical field environment parameters to obtain a second optical field environment parameter; and performing data standardization processing on the second optical field environment parameters to obtain a target optical field environment parameter set.

[0030] The target underwater equipment includes various underwater operation carriers, such as those suitable for deep-sea exploration, underwater robots, submarines, and underwater engineering equipment.

[0031] Optionally, the raw optical field environmental parameters can be acquired in real time using devices such as underwater polarization light sensor arrays, turbidity sensors, and illumination sensors. The raw optical field environmental parameters refer to the unprocessed raw data, which may include, but are not limited to, polarization parameters (degree of polarization, polarization angle, light intensity), water turbidity parameters, and illumination intensity parameters.

[0032] For the three processes in preprocessing, outlier removal can use the 3σ criterion to remove abnormal data caused by sensor failure, noise filtering can use Gaussian filtering to eliminate environmental noise interference, and data standardization can normalize the parameters to the same order of magnitude, which is convenient for subsequent analysis.

[0033] The target light field environment parameter set is standardized data obtained by preprocessing the original light field environment parameters.

[0034] In the specific implementation, firstly, underwater light field environmental parameters are collected in real time using devices such as underwater polarization light sensor array, turbidity sensor, and light sensor; then, the collected light field environmental parameters are preprocessed to obtain a preprocessed light field environmental parameter set.

[0035] Step S102: Based on the target light field environment parameter set, perform light field scene recognition to obtain the current light field scene type; In some embodiments, step S102 may include: performing feature extraction processing on the target light field environment parameter set to obtain the current light field environment features; wherein, the current light field environment features include the current polarization degree variance, the current polarization angle fluctuation amplitude, the current light intensity change rate, and the current turbidity value; using a cosine similarity algorithm to calculate the global similarity between the current light field environment features and the standard scene features in the preset standard scene feature library; if the global similarity is greater than or equal to the preset similarity threshold of the target standard scene, then the target standard scene is determined as the current light field scene type.

[0036] Optionally, light field scene recognition mainly includes two processes: feature extraction and scene matching. Feature extraction refers to extracting four core features from the target light field environment parameter set: polarization variance (reflecting the stability of the polarized light signal), polarization angle fluctuation amplitude (reflecting the drastic change in polarization angle), light intensity change rate (reflecting the speed of change in light intensity), and turbidity value; that is, extracting the current light field environment features. Scene matching refers to comparing the extracted four core features with standard scene features in a preset standard scene feature library to match the current light field scene type.

[0037] The light field scene types include normal light field scenes, weak light scenes (light intensity < preset similarity threshold 1), high turbidity scenes (turbidity > preset similarity threshold 2), and light intensity abrupt change scenes (light intensity change rate > preset similarity threshold 3). For the preset standard scene feature library, the judgment thresholds for each scene are: preset similarity threshold 1 (weak light scene) = 50 lux, preset similarity threshold 2 (high turbidity scene) = 50 NTU, and preset similarity threshold 3 (light intensity abrupt change scene) = 50 lux / s. The preset standard scene feature library contains the standard range of the core features (polarization variance, polarization angle fluctuation amplitude, light intensity change rate, and turbidity value) for each scene.

[0038] Global similarity refers to the comprehensive matching value obtained by taking the four core features extracted—polarization degree variance, polarization angle fluctuation amplitude, light intensity change rate, and turbidity value—as a set of feature vectors and performing cosine similarity calculation with the complete feature vectors in the preset standard scene feature library. It represents the overall similarity between the current set of light field features and a certain type of standard scene. When the overall similarity reaches 85% or above, it is determined to belong to the corresponding light field scene.

[0039] In the specific implementation, the current underwater light field scene type is identified by feature extraction (extracting four core features from the preprocessed light field environment parameter set, namely, the polarization degree variance, the polarization angle fluctuation amplitude, the light intensity change rate, and the turbidity value) and scene matching (comparing the four core features with the standard scene features in the preset standard scene feature library).

[0040] Step S103: Based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, dynamically iteratively optimize the benchmark decision parameters corresponding to the current light field scene type to obtain the optimal decision parameters. In some embodiments, step S103 may include: determining the baseline decision parameters corresponding to the current light field scene type based on the current light field scene type; using a particle swarm optimization algorithm, combined with the dynamic changes of the light field environment parameters in the target light field environment parameter set, to dynamically iteratively optimize the baseline decision parameters to obtain the optimal decision parameters that are suitable for the current light field scene type.

[0041] In the specific implementation, a baseline decision parameter library (including signal acquisition threshold, number of solution iterations, filtering coefficients, and multi-source fusion weights) is preset for different light field scenarios. For example, the baseline decision parameters for each light field scenario are as follows: Normal light field scene: signal acquisition threshold = 0.1, number of solution iterations = 50, filter coefficient = 0.2, multi-source fusion weight (polarized light: inertial: underwater acoustic) = 0.7:0.2:0.1; Low-light scenario: Signal acquisition threshold = 0.05, number of solution iterations = 80, filter coefficient = 0.3, multi-source fusion weight = 0.6:0.3:0.1; High turbidity scenario: signal acquisition threshold = 0.08, number of solution iterations = 100, filter coefficient = 0.4, multi-source fusion weight = 0.5:0.3:0.2; Light intensity sudden change scenario: signal acquisition threshold = 0.12, number of solution iterations = 60, filter coefficient = 0.35, multi-source fusion weight = 0.65:0.25:0.1.

[0042] In the specific implementation, firstly, based on the identified light field scene type and combined with the real-time changes of the light field environment parameters in the preprocessed light field environment parameter set, a particle swarm optimization algorithm is used to dynamically iteratively optimize the baseline decision parameters corresponding to the current light field scene, obtaining the optimal decision parameters adapted to the current light field environment. By dynamically optimizing the navigation decision parameters (signal acquisition threshold, number of solution iterations, filtering coefficients, and fusion weights) according to the real-time light field environment, the problems of increased solution errors and navigation failure caused by fixed parameters can be avoided.

[0043] In some embodiments, the method may further include: during the process of light field scene recognition, performing linear fitting on the light field environment parameters continuously collected in the target light field environment parameter set to obtain parameter change trend data; identifying light field anomalies based on the parameter change trend data and a preset parameter change threshold; if it is determined that there are anomalies in the light field environment parameters in the target light field environment parameter set, outputting a light field anomaly warning signal for the current light field scene type; after obtaining the light field anomaly warning signal, automatically adjusting the baseline decision parameters corresponding to the current light field scene type according to a preset anomaly adjustment strategy; and dynamically iteratively optimizing the adjusted baseline decision parameters based on the current light field scene type and the dynamic changes in the light field environment parameters in the target light field environment parameter set to obtain the optimal decision parameters.

[0044] In this embodiment, for light field anomaly scenarios, when an anomaly warning signal is received, the parameter optimization strategy of the baseline decision parameters corresponding to the current light field scenario is automatically adjusted. This includes increasing the filter coefficient and the number of solution iterations to improve signal anti-interference capability and solution accuracy. Finally, a particle swarm optimization algorithm is used to dynamically iteratively optimize the adjusted baseline decision parameters to obtain the optimal decision parameters under the light field anomaly scenario. Specifically, when an anomaly warning signal is received, the current light field scenario type is first determined, and then its corresponding baseline decision parameters are used as the initial particle swarm or search starting point for optimization. Then, the number of solution iterations in the baseline decision parameters is increased by 20%, and the filter coefficient is increased by 10%. Finally, combined with the latest real-time light field environment parameters, particle swarm optimization iteration is performed on the adjusted baseline decision parameters to ultimately search for a set of optimal decision parameters that balance accuracy and anti-interference capability under the current anomaly interference trend.

[0045] Step S104: Determine the optimal solution algorithm based on the current light field scene type, and combine it with the optimal decision parameters to solve the polarization parameters in the target light field environment parameter set to obtain navigation data; In the specific implementation, a multi-navigation solution algorithm library is pre-built. This library includes a Kalman filter algorithm (suitable for normal scenes), an adaptive least squares algorithm (suitable for low-light scenes), an anti-interference particle filter algorithm (suitable for high-turbidity scenes), and a fast solution algorithm (suitable for scenes with sudden changes in light intensity). Specifically, the parameter settings for each algorithm in the multi-navigation solution algorithm library are as follows: Kalman filter algorithm: process noise variance Q=diag([1e-6,1e-6,1e-6]), observation noise variance R=diag([1e-5,1e-5]); the specific values ​​of the process noise variance Q and the observation noise variance R are a set of optimized parameters calibrated through experiments in combination with the application environment of complex underwater optical fields. Adaptive least squares algorithm: Iteration accuracy ε = 1e-8, maximum number of iterations = 100; Anti-interference particle filter algorithm: Particle number = 1000, resampling threshold = 0.5; Fast solution algorithm: Solution step size = 0.01s, accuracy threshold = .

[0046] Specifically, based on the current light field scene type, the algorithm automatically switches to the corresponding solution algorithm: normal scenes switch to the Kalman filter algorithm, low-light scenes switch to the adaptive least squares algorithm, high-turbidity scenes switch to the anti-interference particle filter algorithm, and scenes with sudden changes in light intensity switch to the fast solution algorithm.

[0047] In this embodiment, firstly, based on the current light field scene type, the system automatically switches to the optimal solution algorithm and calls the optimal decision parameters to solve the preprocessed polarized light parameters, obtaining navigation data such as the carrier's heading angle, attitude angle, and positioning coordinates. By constructing a multi-algorithm adaptive switching mechanism, the optimal solution algorithm is selected for computation according to different light field scenes, effectively reducing the solution deviation under complex light fields and significantly improving the accuracy and computational efficiency of navigation solutions in complex underwater environments. Step S105: Verify the navigation data and output a navigation decision command that has passed the verification.

[0048] In some embodiments, step S105 may include: fusing and verifying the navigation data with the inertial navigation data and underwater acoustic navigation auxiliary data corresponding to the target underwater device to obtain the current polarization navigation error; if the current polarization navigation error is less than or equal to a preset navigation accuracy threshold, then the navigation data verification is determined to be successful, and a navigation decision command is generated based on the verified navigation data; if the current polarization navigation error is greater than the preset navigation accuracy threshold, then the navigation data is discarded, and the process returns to the step of determining the optimal solution algorithm based on the current light field scene type, and combining the optimal decision parameters to solve the polarization light parameters in the target light field environment parameter set to obtain the navigation data, until a verified navigation decision command is output.

[0049] In the specific implementation, the calculated navigation data is verified (by fusing and verifying the navigation data with the inertial navigation and underwater acoustic navigation auxiliary data of the underwater equipment, and eliminating data with excessive calculation errors). After successful verification, precise navigation decision commands (including navigation parameters, navigation accuracy assessment, and light field environment early warning information) are output. Simultaneously, the navigation accuracy data and light field environment parameters are fed back to the decision parameter optimization step and the algorithm switching strategy step, forming a closed-loop optimization mechanism to continuously optimize the decision parameters and algorithm switching strategy, thereby improving the long-term stability and accuracy of the navigation system.

[0050] Specifically, when the current polarization navigation error of the navigation data calculated in step S104 exceeds the preset navigation accuracy threshold, the abnormal navigation data is discarded, and the process returns to step S104 first. Based on the preprocessed light field environment parameter set and optimal decision parameters obtained at the current moment, the navigation calculation is re-executed to obtain new navigation data with minimal delay. If the navigation data error after recalculation in step S104 exceeds the preset navigation accuracy threshold multiple times consecutively (e.g., 3 times consecutively, the number of times can be set according to the actual situation), the overall data quality at the current moment is determined to be unreliable. In this case, the process returns to step S101, the original light field environment parameters are re-acquired, and the complete data preprocessing, scene recognition, parameter optimization, and calculation process is executed until valid navigation data is obtained. Through the above two-level retry mechanism, the real-time nature of navigation decision-making is guaranteed, and self-recovery capability is provided when data quality is severely degraded.

[0051] Steps S101 to S105 as illustrated in this embodiment of the application complete accurate light field scene recognition based on the target light field environment parameter set. This effectively distinguishes different underwater light field conditions, enabling intelligent perception and discrimination of complex underwater light field environments. It provides accurate scene-based information for subsequent parameter adaptive optimization and algorithm matching switching. Based on the identified current light field scene type and combined with the real-time dynamic changes in light field environment parameters, the corresponding baseline decision parameters are dynamically iteratively optimized. This adapts to the real-time changing characteristics of different underwater light field environments, overcoming the shortcomings of traditional fixed baseline parameters that cannot adapt to complex dynamic underwater scenes, and ensuring accuracy under different operating conditions. It can obtain optimal navigation decision parameters for various scenarios, significantly improving the adaptability and accuracy of navigation environments. It matches the optimal solution algorithm based on the type of light field scene and combines the optimized decision parameters to complete the polarization parameter calculation. This allows for targeted adaptation to the calculation needs of different light field scenes, significantly improving the calculation accuracy and output stability of underwater polarization navigation. By verifying the calculated navigation data and filtering out compliant and valid navigation decision commands, it effectively avoids navigation deviations caused by abnormal calculation data output, greatly improving the accuracy and effectiveness of underwater navigation decisions, thereby ensuring the precision and reliability of underwater equipment navigation operations.

[0052] To explain in detail the principles of the technical solution of this application, the overall process of this application will be described below with reference to some specific embodiments. It is easy to understand that the following is an explanation of the technical principles of this application and should not be regarded as a limitation of this application.

[0053] The purpose of this application is to address the technical problems of current polarization navigation software decision-making methods and systems, such as their inability to adapt to complex underwater optical fields, low navigation accuracy, poor stability, slow decision response, and insufficient reliability. This application provides an adaptive polarization navigation software decision-making method and system for complex underwater optical fields, enabling real-time perception of the underwater optical field environment, dynamic optimization of decision parameters, and adaptive switching of solution algorithms. This improves the accuracy, stability, and reliability of polarization navigation in complex underwater scenarios, meeting the high-precision autonomous navigation needs of various underwater vehicles.

[0054] The adaptive polarization navigation decision-making method provided in this application is a polarization navigation software decision-making method that can perceive the underwater light field environment in real time, dynamically optimize decision parameters and calculation algorithms, and has high reliability and adaptability. It belongs to the fields of underwater navigation technology, polarized light application and software decision-making technology, and is suitable for high-precision autonomous navigation scenarios of various underwater vehicles such as deep-sea exploration, underwater robots, submarines, and underwater engineering equipment. It is especially suitable for complex environments such as uneven underwater light field, water turbidity fluctuations, and sudden changes in light intensity.

[0055] Please see Figure 2 , Figure 2This is a block diagram of the module structure of an adaptive polarization navigation software decision system for complex underwater optical fields, provided in an embodiment of this application. Figure 2 This is the system architecture corresponding to an adaptive polarization navigation decision-making method provided in the embodiments of this application, such as... Figure 2 As shown, the parameter acquisition module collects raw light field environmental parameters. These parameters are then fed into the data preprocessing module for outlier removal, noise reduction, and standardization. The navigation parameter calculation module performs preliminary calculations of basic physical quantities. Simultaneously, the external interface module and self-check / maintenance module handle data interaction between the system and the outside world, as well as pre-start status calibration. The decision logic generation module receives refined features from the data preprocessing module. Internally, it integrates scene recognition, parameter optimization, algorithm switching, and solution processing. By analyzing the current light field scene type, it adaptively optimizes the decision parameters and calls the optimal solution algorithm for final calculation. The solution results are then aggregated at the decision center and distributed to the following four modules for collaborative work: Figure 2 The bottom left decision output module formally issues heading and attitude control commands to the underwater vehicle; the data fusion processing module introduces inertial navigation and underwater acoustic data for verification, achieving multi-source fusion; the environmental adaptability adjustment module feeds back the calculation accuracy to the decision center, achieving closed-loop parameter correction; the system status monitoring module monitors the operating status in real time, and triggers the fault-tolerant backup logic of the fault-tolerant backup module upon detecting anomalies. Furthermore... Figure 2 The decision output module on the right serves as execution confirmation feedback, ensuring accurate and closed-loop instruction execution.

[0056] Please see Figure 3 , Figure 3 This is a flowchart illustrating an adaptive polarization navigation software decision-making method for complex underwater optical fields provided in an embodiment of this application. Figure 3 As shown, the overall implementation process of an adaptive polarization navigation decision method provided in this application embodiment may include, but is not limited to, steps S301 to S305.

[0057] Step S301: Real-time acquisition and preprocessing of light field environmental parameters; In the specific implementation, firstly, underwater light field environmental parameters are collected in real time using devices such as underwater polarization light sensor arrays, turbidity sensors, and illumination sensors. These parameters include polarization light parameters (degree of polarization, polarization angle, and light intensity), water turbidity parameters, and illumination intensity parameters. Then, the collected light field environmental parameters are preprocessed to obtain a preprocessed set of light field environmental parameters. The preprocessing includes outlier removal (using the 3σ criterion to remove abnormal data caused by sensor malfunctions), noise filtering (using Gaussian filtering to eliminate environmental noise interference), and data standardization (normalizing the parameters to the same order of magnitude for easier subsequent analysis).

[0058] Step S302, light field scene recognition and anomaly prediction; Please see Figure 4 , Figure 4 This is a schematic diagram of a light field scene recognition process provided in an embodiment of this application, such as... Figure 4 As shown, the "feature extraction unit" acquires core features such as underwater light field environmental parameters through light field data acquisition and feature point extraction; the "scene matching unit" uses database retrieval and similarity calculation to determine the current light field scene type (such as a high turbidity scene); at the same time, the "anomaly prediction unit" performs threshold comparison and anomaly marking simultaneously, and outputs an anomaly warning signal if the trend of consecutive frames exceeds the limit.

[0059] In the specific implementation, firstly, through feature extraction (extracting four core features from the preprocessed light field environment parameter set: polarization degree variance, polarization angle fluctuation amplitude, light intensity change rate, and turbidity value) and scene matching (comparing the similarity of the four core features with the standard scene features in the preset standard scene feature library), the current underwater light field scene type is identified. The light field scene type includes normal light field scene, weak light scene (light intensity < preset similarity threshold 1), high turbidity scene (turbidity > preset similarity threshold 2), and light intensity change scene (light intensity change rate > preset similarity threshold 3). At the same time, through trend analysis (linear fitting of continuously collected parameters from the preprocessed light field environment parameter set to predict parameter change trends), light field anomalies (such as polarization signal distortion, sudden drop / rise in light intensity, and sudden turbidity change) are identified, and anomaly warning signals are output.

[0060] Step S303: Adaptive optimization of decision parameters; Please see Figure 5 , Figure 5 This is a flowchart illustrating an adaptive optimization process for decision parameters provided in an embodiment of this application, such as... Figure 5 As shown, the "benchmark parameter call unit" extracts the initial decision parameter set from the benchmark decision parameter library according to the light field scene type identified by S302; the "parameter optimization unit" runs the particle swarm optimization algorithm and performs algorithmic iterative adjustments to the initial decision parameter set based on real-time environmental feedback to ensure that the parameters achieve optimal fitness in the current light field scene; finally, after confirmation by the "optimal parameter output unit", the final result is sent to the solution module.

[0061] In the specific implementation, a baseline decision parameter library (including signal acquisition threshold, number of solution iterations, filtering coefficients, multi-source fusion weights, etc.) is preset for different light field scenarios. First, based on the light field scenario type identified in step S302, and combined with the real-time changes of the light field environment parameters in the preprocessed light field environment parameter set in step S301, a particle swarm optimization algorithm is used to dynamically iteratively optimize the baseline decision parameters corresponding to the current light field scenario to obtain the optimal decision parameters adapted to the current light field environment. Furthermore, for abnormal light field scenarios, when an abnormal warning signal is received from step S302, the parameter optimization strategy of the baseline decision parameters corresponding to the current light field scenario is automatically adjusted, such as increasing the filtering coefficient and the number of solution iterations to improve signal anti-interference capability and solution accuracy. Finally, a particle swarm optimization algorithm is used again to dynamically iteratively optimize the adjusted baseline decision parameters to obtain the optimal decision parameters under abnormal light field scenarios.

[0062] Step S304: Adaptive switching of multiple algorithms and navigation solution; Please see Figure 6 , Figure 6 This is a logical diagram illustrating a multi-algorithm adaptive switching method provided in an embodiment of this application, such as... Figure 6 As shown, the "algorithm library storage unit" contains multiple pre-set collections designed for different light field scenarios (such as anti-interference particle filter algorithms); the "algorithm selection unit" receives the real-time environmental parameter input after preprocessing in S301 and the light field scenario type output in step S302, and automatically retrieves the most matching algorithm for the current light field scenario based on the matching algorithm decision logic; the selected algorithm runs in the "algorithm execution unit", and is solved in combination with the optimal decision parameters input in S303, and finally outputs the navigation data results.

[0063] In the specific implementation, a multi-navigation solution algorithm library is constructed, which includes the Kalman filter algorithm (suitable for normal scenarios), the adaptive least squares algorithm (suitable for low-light scenarios), the anti-interference particle filter algorithm (suitable for high-turbidity scenarios), and the fast solution algorithm (suitable for scenarios with sudden changes in light intensity). First, based on the light field scenario type identified in step S302, the optimal solution algorithm is automatically switched to, and the optimal decision parameters obtained in step S303 are called to solve the polarization light parameters after preprocessing in step S301, so as to obtain navigation data such as the heading angle, attitude angle, and positioning coordinates of the carrier.

[0064] Step S305: Decision output and closed-loop optimization.

[0065] In the specific implementation, the navigation data obtained in step S304 is verified (the navigation data is fused and verified with the inertial navigation and underwater acoustic navigation auxiliary data of the underwater equipment, and data with excessive calculation error is removed). After the verification is passed, accurate navigation decision instructions (including navigation parameters, navigation accuracy assessment, and light field environment early warning information) are output. At the same time, the navigation accuracy data and light field environment parameters are fed back to steps S303 and S304 to form a closed-loop optimization mechanism, continuously optimize the decision parameters and algorithm switching strategy, and improve the long-term stability and accuracy of the navigation system.

[0066] It should be noted that this embodiment is only a brief illustrative description of the overall process of an adaptive polarization navigation decision method. Detailed descriptions of each step can be found in the relevant content of the foregoing embodiments, and will not be repeated here. It is understood that this application does not impose any limitations on this.

[0067] The solutions of this application embodiment will be described in detail and explained below with reference to specific application examples: In this embodiment, the adaptive polarization navigation software decision-making method for complex underwater optical fields is applied to a deep-sea exploration underwater robot. The underwater working depth is 500-800 meters, the water turbidity fluctuates between 20-80 NTU, and the light intensity changes significantly with depth, presenting complex scenarios such as sudden changes in light intensity and uneven turbidity. The specific implementation steps are as follows: Steps A-E: Specifically, the system modules corresponding to the adaptive polarization navigation decision-making method provided in this application mainly include a parameter acquisition module, a data preprocessing module, a scene recognition module, a parameter optimization module, an algorithm switching module, a solution module, a decision output module, and a fault-tolerant backup module. Steps A-E below are specifically implemented through these system modules.

[0068] Step A: Real-time acquisition and preprocessing of light field environmental parameters; In the specific implementation, a parameter acquisition module is used to acquire real-time optical field environmental parameters. This module consists of an underwater polarization sensor array, a turbidity sensor, and an illumination sensor. It is used to acquire underwater optical field environmental parameters (polarization degree, polarization angle, light intensity, turbidity, and illumination intensity) in real time and transmit the acquired raw data to a data preprocessing module. Specifically, the polarization sensor array uses a 4-channel CMOS polarization sensor (model "XXX"), operating at 3.3V, with a power consumption ≤50mW, an IP68 underwater protection rating, and stable operation at depths of 0-1000 meters. The polarizer angles are 0°, 45°, 90°, and 135°, and the acquisition frequency is adjustable (50-200Hz). The turbidity sensor uses a fiber optic turbidity sensor (model "XXX") with a measurement range of 0-100 NTU and an accuracy of ±1 NTU. The illumination sensor uses an underwater low-light sensor (model "XXX") with a measurement range of 0-500 lux and an accuracy of ±0.1 lux. Each sensor is connected to the data preprocessing module via an SPI interface, transmitting the acquired raw data to the data preprocessing module in real time at a rate of 1 Mbps.

[0069] The data preprocessing module connects to the parameter acquisition module and receives the raw data transmitted from it. It performs preprocessing operations such as outlier removal, noise filtering, and data standardization to remove interference and anomalies from the data, obtaining a standardized set of light field environment parameters, which is then transmitted to the scene recognition module. The data preprocessing module uses a microcontroller as its core processor with a clock speed of 480MHz and 1MB of built-in SRAM for performing data preprocessing operations. The software is programmed in C language to implement functions such as 3σ outlier removal, Gaussian filtering, and min-max normalization. The preprocessing latency is ≤10ms, and the preprocessed data is transmitted to the scene recognition module via a CAN bus.

[0070] Specifically, a 4-channel polarization light sensor array (polarizer angles of 0°, 45°, 90°, and 135°), an optical fiber turbidity sensor, and an underwater low-light sensor are used to collect underwater light field environmental parameters in real time at a frequency of 100Hz. The polarization light parameter acquisition range is: polarization degree 0-1, polarization angle 0-360°, and light intensity 0-1000 lux; the turbidity parameter acquisition range is 0-100 NTU; and the light intensity acquisition range is 0-500 lux.

[0071] The collected raw optical field environmental parameters are preprocessed as follows: First, the 3σ criterion is used to remove outliers (such as polarization greater than 1 or negative turbidity caused by sensor failure). Then, Gaussian filtering (with a filter window size of 5) is applied to the parameters obtained after outlier removal to eliminate environmental noise interference and reduce data fluctuations. Finally, the min-max normalization method is used to standardize all parameters after noise filtering to the [0,1] interval to obtain the preprocessed optical field environmental parameter set.

[0072] Step B: Light field scene recognition and anomaly prediction; In its implementation, a scene recognition module is used to identify and predict anomalies in the light field scene. This module is connected to the data preprocessing module and includes a built-in light field scene recognition model and a standard scene feature library. It receives the preprocessed light field environment parameter set, extracts core features, matches them with the standard scene feature library, and identifies the current light field scene type. Simultaneously, through parameter trend analysis, it identifies light field anomalies and outputs warning signals. The scene recognition results and anomaly warning signals are transmitted to the parameter optimization module and the algorithm switching module. The scene recognition module uses an FPGA chip as its core processor and includes a built-in light field scene recognition model and a standard scene feature library. The standard scene feature library is stored in the FPGA's built-in Flash memory and can be flexibly updated via a host computer. Feature extraction and cosine similarity comparison are implemented through hardware logic, achieving a scene recognition latency of ≤20ms and an accuracy rate of ≥95%. The scene recognition results and anomaly warning signals are transmitted to the parameter optimization module via an Ethernet interface. Similarly, the scene recognition results can be transmitted to the algorithm switching module via an Ethernet interface.

[0073] The light field scene recognition model can employ a deep learning-based neural network model, such as a lightweight convolutional neural network (e.g., MobileNet V3, ShuffleNet V2), a recurrent neural network (Long Short-Term Memory, LSTM), or a Transformer network based on a self-attention mechanism. Taking MobileNet V3 as an example, the preprocessed target light field environment parameter set is reorganized into a multi-dimensional feature map input network. Its depthwise separable convolutional layers are used for rapid feature extraction and dimensionality reduction. Finally, a fully connected layer, in conjunction with a Softmax classifier, outputs four core feature values ​​(polarization variance, polarization angle fluctuation amplitude, light intensity change rate, and turbidity value) or directly outputs the probability distribution of the scene type. Using such a lightweight network can significantly reduce the computational load on the underwater embedded processor while ensuring recognition accuracy, meeting the millisecond-level real-time requirements of the navigation system. Furthermore, those skilled in the art can also use traditional machine learning classifiers such as Support Vector Machine (SVM) or Random Forest to achieve scene recognition; this application does not limit this approach.

[0074] In practical applications, a standard scene feature library is pre-built, in which the judgment thresholds for each scene are: preset similarity threshold 1 (weak light scene) = 50 lux, preset similarity threshold 2 (high turbidity scene) = 50 NTU, and preset similarity threshold 3 (light intensity change scene) = 50 lux / s. The standard scene feature library contains the standard range of the core features of each scene (polarization variance, polarization angle fluctuation amplitude, light intensity change rate, and turbidity value).

[0075] Specifically, firstly, the light field scene recognition model extracts four core features from the light field environment parameter set after preprocessing in step A: polarization variance (reflecting the stability of polarized light signals), polarization angle fluctuation amplitude (reflecting the drastic change in polarization angle), light intensity change rate (reflecting the rate of change in light intensity), and turbidity value. Then, the cosine similarity algorithm is used to compare the four extracted core features with features in the standard scene feature library. When the similarity is ≥85%, it is determined to be the corresponding scene.

[0076] Simultaneously, linear fitting is performed on the parameters collected from 10 consecutive frames of the light field environment parameter set after preprocessing in step A to predict the parameter change trend: when the predicted turbidity will exceed 50 NTU, the light intensity will be lower than 50 lux, or the light intensity change rate will exceed 50 lux / s, an abnormal light field warning signal is output to prompt subsequent modules to adjust the decision parameter optimization strategy.

[0077] Step C: Adaptive optimization of decision parameters; In the specific implementation, an adaptive optimization module for decision parameters is used. This module is connected to both the scene recognition module and the algorithm switching module. It incorporates a baseline decision parameter library and a particle swarm optimization algorithm. The module receives scene type and anomaly warning signals from the scene recognition module, calls the corresponding parameters from the baseline decision parameter library, and performs dynamic iterative optimization using the particle swarm optimization algorithm to obtain the optimal decision parameters. These optimal decision parameters are then transmitted to the algorithm switching module and the solution module. The parameter optimization module uses an embedded processor and incorporates the baseline decision parameter library and particle swarm optimization algorithm. The baseline decision parameter library supports offline configuration and online updates. The algorithm is implemented in Python, runs at 100Hz, and has a parameter optimization latency of ≤30ms. The optimal decision parameters are transmitted to the algorithm switching module and the solution module via a PCIe interface.

[0078] In practical applications, a baseline decision parameter library is pre-built, with the following baseline decision parameters for each scenario: Normal light field scene: signal acquisition threshold = 0.1, number of solution iterations = 50, filter coefficient = 0.2, multi-source fusion weight (polarized light: inertial: underwater acoustic) = 0.7:0.2:0.1; Low-light scenario: Signal acquisition threshold = 0.05, number of solution iterations = 80, filter coefficient = 0.3, multi-source fusion weight = 0.6:0.3:0.1; High turbidity scenario: signal acquisition threshold = 0.08, number of solution iterations = 100, filter coefficient = 0.4, multi-source fusion weight = 0.5:0.3:0.2; Light intensity sudden change scenario: signal acquisition threshold = 0.12, number of solution iterations = 60, filter coefficient = 0.35, multi-source fusion weight = 0.65:0.25:0.1.

[0079] Specifically, a particle swarm optimization algorithm is used to dynamically optimize the baseline decision parameters. The particle swarm size is set to 30, the number of iterations is set to 20, and the fitness function is set to minimize the navigation accuracy error (target error ≤ 0.05°). Based on the current light field scene type identified in step B, the corresponding baseline decision parameters are called and combined with the preprocessed real-time light field environment parameters. The optimal decision parameters are obtained through iterative optimization using the particle swarm optimization algorithm. For example, when the scene is identified as a high-turbidity scene (turbidity = 60 NTU), the optimized parameters are: signal acquisition threshold = 0.07, number of iterations = 105, filter coefficient = 0.42, and multi-source fusion weight = 0.48:0.32:0.2.

[0080] In addition, when the light field anomaly warning signal output in step B is received, the optimization strategy of the corresponding benchmark decision parameters is automatically adjusted, such as increasing the number of solution iterations by 20% and the filtering coefficient by 10% to improve the anti-interference capability. Finally, the particle swarm optimization algorithm is used to dynamically iterate and optimize the adjusted benchmark decision parameters to obtain the optimal decision parameters under the light field anomaly scenario.

[0081] Step D: Adaptive switching between multiple algorithms and navigation solution; In the specific implementation, an algorithm switching module enables adaptive switching between multiple algorithms, while a solution module performs navigation solution. The algorithm switching module is connected to the scene recognition module, parameter optimization module, and solution module, and has a built-in multi-navigation solution algorithm library. It receives the scene type from the scene recognition module, automatically switches to the optimal solution algorithm, and calls the optimal decision parameters from the parameter optimization module, transmitting the algorithm selection command and optimal decision parameters to the solution module. The solution module, connected to the algorithm switching module, receives the algorithm selection command and optimal decision parameters from the algorithm switching module, performs navigation solution on the preprocessed polarized light parameters, obtains navigation data such as the carrier's heading angle, attitude angle, and positioning coordinates, and transmits the navigation data to the decision output module. Specifically, the algorithm switching module and parameter optimization module share an embedded processor and have a built-in multi-navigation solution algorithm library that supports dynamic addition and updates. Automatic algorithm switching is achieved through software logic, with a switching latency of ≤10ms. Algorithm selection commands and optimal decision parameters are synchronously transmitted to the solution module. The calculation module uses a DSP chip with a main frequency of 456MHz to execute navigation calculation algorithms. The calculation frequency is 100Hz and the calculation delay is ≤20ms. The calculation data is transmitted to the decision output module via CAN bus.

[0082] In practical applications, a multi-navigation solution algorithm library is pre-built, and the parameter settings for each algorithm are as follows: Kalman filter algorithm: process noise variance Q=diag([1e-6,1e-6,1e-6]), observation noise variance R=diag([1e-5,1e-5]); Adaptive least squares algorithm: Iteration accuracy ε = 1e-8, maximum number of iterations = 100; Anti-interference particle filter algorithm: Particle number = 1000, resampling threshold = 0.5; Fast solution algorithm: Solution step size = 0.01s, accuracy threshold = .

[0083] Specifically, based on the current light field scene type identified in step B, the corresponding solution algorithm is automatically switched: normal scene switches to Kalman filter algorithm, weak light scene switches to adaptive least squares algorithm, high turbidity scene switches to anti-interference particle filter algorithm, and light intensity change scene switches to fast solution algorithm; then the optimal decision parameters obtained in step C are called to solve the preprocessed polarized light parameters to obtain the underwater robot's heading angle, pitch angle, roll angle and positioning coordinates.

[0084] Step E: Decision output and closed-loop optimization.

[0085] In the specific implementation, a decision output module is used to realize decision output and closed-loop optimization. The decision output module is connected to the solution module and is used to receive navigation data transmitted by the solution module. It performs fusion verification by combining inertial navigation and underwater acoustic navigation auxiliary data. After the verification is successful, it outputs accurate navigation decision commands. At the same time, it feeds back navigation accuracy data and optical field environment parameters to the parameter optimization module and algorithm switching module to achieve closed-loop optimization. The decision output module is implemented using a microcontroller. It receives navigation data transmitted by the solution module and performs fusion verification by combining inertial navigation and underwater acoustic navigation auxiliary data (received via RS232 interface). The verification delay is ≤10ms. After the verification is successful, it outputs navigation decision commands through the RS485 interface. At the same time, it transmits feedback data to the parameter optimization module and algorithm switching module through the Ethernet interface to achieve closed-loop optimization.

[0086] Specifically, the navigation data obtained in step D is fused and verified with the underwater robot's built-in inertial navigation and underwater acoustic navigation auxiliary data. A navigation accuracy threshold is set (heading angle error ≤ 0.05°, positioning error ≤ 0.5m). When the error of the navigation data obtained in step D exceeds the threshold, the abnormal navigation data is discarded, and the process returns to step D first. Based on the preprocessed light field environment parameter set and optimal decision parameters obtained at the current moment, the navigation calculation is re-executed to obtain new navigation data with minimal delay. If the error of the navigation data after recalculation exceeds the threshold multiple times (e.g., 3 times consecutively, the number of times can be set according to the actual situation), the overall data quality at the current moment is determined to be unreliable. At this time, the process returns to step A, the original light field environment parameters are re-acquired, and the complete data preprocessing, scene recognition, parameter optimization, and calculation process is executed until verified navigation data is obtained. Through the above two-level retry mechanism, the real-time nature of navigation decision-making is guaranteed, and self-recovery capability is provided when the data quality is severely degraded.

[0087] After the navigation data obtained in step D passes verification, navigation decision commands are output via the RS485 interface, including heading angle, attitude angle, positioning coordinates, navigation accuracy assessment (e.g., "Currently in a high turbidity scene, navigation accuracy 0.04°, light field stable"), and light field environment warning information (e.g., "Turbidity is predicted to increase; please pay attention to navigation accuracy"). Simultaneously, navigation accuracy data (e.g., actual calculation error) and real-time light field environment parameters are fed back to steps C and D, dynamically adjusting the initial values ​​of the baseline decision parameters and the algorithm's parameter settings to form a closed-loop optimization, continuously improving navigation accuracy and stability. For example, when the navigation accuracy error is lower than [a certain value] for 5 consecutive frames... In such cases, appropriately reducing the number of solution iterations for the corresponding scenario can improve decision-making efficiency.

[0088] In steps A-E of this embodiment, the overall power consumption of the system is ≤5W, the operating temperature range is -10℃ to 60℃, the underwater protection level is IP68, and it can work continuously and stably in complex light field environments with a depth of 500-800 meters and a turbidity of 20-80 NTU. The navigation accuracy meets the actual needs of underwater robots for deep-sea exploration.

[0089] In addition, the adaptive polarization navigation decision system also includes a fault-tolerant backup module. This module is connected to each module of the system and is used to monitor the operating status of each module in real time. When a module fails (such as sensor failure, data transmission failure, or calculation failure), it automatically switches to the backup module, invoking preset emergency decision parameters and backup algorithms to ensure the continuous and stable operation of the navigation system and avoid navigation interruptions. The fault-tolerant backup module adopts a dual-redundancy design, with backup modules for each core module (parameter acquisition module, data preprocessing module, and calculation module). The backup modules operate synchronously with the main modules. A fault monitoring circuit monitors the operating status of each module in real time (such as power supply status, data transmission status, and algorithm running status). When a failure of the main module is detected, it automatically switches to the backup module with a switching delay of ≤50ms, ensuring the continuous and stable operation of the navigation system.

[0090] In summary, the adaptive polarization navigation decision-making method provided in this application, compared with related technologies, has the following characteristics and effects: (1) This application realizes the automatic identification of light field scene and the early prediction of light field anomalies by collecting underwater light field environmental parameters in real time. It dynamically optimizes decision parameters and adaptively switches calculation algorithms according to different scenes (normal, weak light, high turbidity, light intensity change). It solves the defects of fixed parameters and single algorithm in related technologies that cannot adapt to complex light fields. It can maintain high navigation accuracy in various complex underwater scenes.

[0091] (2) This application constructs a multi-algorithm adaptive switching mechanism and a decision parameter closed-loop optimization system. Combined with an improved anti-interference solution algorithm and dynamically optimized decision parameters, it effectively suppresses the solution error caused by underwater light field scattering, noise, interference and other factors, and greatly improves the accuracy of heading angle solution and positioning solution. Compared with the current polarization navigation software decision method, this application can not only greatly improve the overall solution accuracy, but also significantly reduce the navigation accuracy fluctuation range, and the system operation stability is also significantly enhanced.

[0092] (3) This application realizes the integrated collaborative design of parameter acquisition, environmental perception, parameter optimization, algorithm calculation and decision output, optimizes the data transmission process and reduces the decision response time; at the same time, it sets up a fault-tolerant backup module and a fault self-diagnosis mechanism. When a certain module of the system is abnormal, it can automatically switch to backup mode to avoid navigation interruption. The system reliability is improved by more than 50%, which can meet the needs of long-term and continuous underwater navigation.

[0093] (4) This application does not require large-scale modification of existing underwater polarization navigation hardware equipment. It can be directly adapted to different types of underwater vehicles (underwater robots, submarines, exploration equipment, etc.) and underwater scenarios with different depths and turbidities. Moreover, the standard scenario feature library, benchmark decision parameters, and solution algorithms can be flexibly adjusted according to actual application needs. It has strong versatility and outstanding practicality. Furthermore, this application is based on the principle of polarization passive navigation and does not rely on external signals (such as satellites or underwater acoustic signals). It can effectively avoid electromagnetic interference and acoustic channel interference. The software decision-making process has low power consumption. Compared with the current navigation system, the overall operating power consumption of the navigation system is significantly reduced. It is suitable for scenarios such as deep sea where there is no external signal coverage and high anti-interference requirements.

[0094] Please see Figure 7 This application also provides an adaptive polarization navigation decision device 700, which can implement the above-described method. The device includes the following modules: The data acquisition and preprocessing module 701 is used to acquire the original light field environment parameters of the underwater environment in which the target underwater device is located at the current moment, and to preprocess the original light field environment parameters to obtain the target light field environment parameter set. The light field scene recognition module 702 is used to perform light field scene recognition based on the target light field environment parameter set to obtain the current light field scene type; The decision parameter optimization module 703 is used to dynamically iteratively optimize the baseline decision parameters corresponding to the current light field scene type based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, so as to obtain the optimal decision parameters. The polarization parameter calculation module 704 is used to determine the optimal calculation algorithm according to the current light field scene type, and combine the optimal decision parameters to calculate the polarization parameters in the target light field environment parameter set to obtain navigation data. The navigation data verification module 705 is used to verify the navigation data and output navigation decision instructions that have passed the verification.

[0095] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0096] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0097] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0098] Please see Figure 8 , Figure 8 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 801 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 802 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called and executed by the processor 801 using the methods described in the embodiments of this application. The 803 input / output interface is used to implement information input and output. The communication interface 804 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 805 transmits information between various components of the device (e.g., processor 801, memory 802, input / output interface 803, and communication interface 804); The processor 801, memory 802, input / output interface 803, and communication interface 804 are connected to each other within the device via bus 805.

[0099] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0100] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0101] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0102] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0103] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0104] This application provides an adaptive polarization navigation decision-making method and related equipment. Based on a target light field environment parameter set, it achieves accurate light field scene recognition, effectively distinguishing different underwater light field conditions and realizing intelligent perception and discrimination of complex underwater light field environments. This provides accurate scene-based information for subsequent parameter adaptive optimization and algorithm matching switching. Based on the identified current light field scene type and combined with the real-time dynamic changes in light field environment parameters, it dynamically iterates and optimizes the corresponding benchmark decision parameters for the scene. This adapts to the real-time changing characteristics of different underwater light field environments, overcoming the shortcomings of traditional fixed benchmark parameters that cannot adapt to complex dynamic underwater scenes, ensuring accurate polarization navigation. Under the same operating conditions, it can obtain navigation decision parameters with optimal adaptability, significantly improving the adaptability of navigation environment and parameter accuracy; it matches the optimal solution algorithm according to the type of light field scene, and completes the polarization parameter solution by combining the optimized optimal decision parameters, which can be specifically adapted to the solution requirements of different light field scenes, significantly improving the solution accuracy and output stability of underwater polarization navigation; by verifying the navigation data obtained by solution, it filters out compliant and valid navigation decision commands, effectively avoiding navigation deviation problems caused by abnormal solution data output, greatly improving the accuracy and effectiveness of underwater navigation decision, and thus ensuring the accuracy and reliability of underwater equipment navigation operations.

[0105] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0106] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0107] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0109] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0110] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0112] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0113] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0114] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. An adaptive polarization navigation decision-making method, characterized in that, The method includes the following steps: The original optical field environment parameters of the underwater environment in which the target underwater device is located at the current moment are obtained, and the original optical field environment parameters are preprocessed to obtain the target optical field environment parameter set. Based on the target light field environment parameter set, light field scene recognition is performed to obtain the current light field scene type; Based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, the baseline decision parameters corresponding to the current light field scene type are dynamically iteratively optimized to obtain the optimal decision parameters. The optimal solution algorithm is determined based on the current light field scene type, and the polarization parameters in the target light field environment parameter set are solved in combination with the optimal decision parameters to obtain navigation data. The navigation data is verified, and a navigation decision command that passes the verification is output.

2. The method according to claim 1, characterized in that, The preprocessing of the original light field environment parameters to obtain the target light field environment parameter set includes: The original optical field environment parameters are subjected to outlier removal processing to obtain the first optical field environment parameters; The first optical field environmental parameters are subjected to noise filtering to obtain the second optical field environmental parameters; The second light field environment parameters are subjected to data standardization processing to obtain the target light field environment parameter set.

3. The method according to claim 1, characterized in that, The process of identifying the light field scene based on the target light field environment parameter set to obtain the current light field scene type includes: The target light field environment parameter set is subjected to feature extraction processing to obtain the current light field environment features; wherein, the current light field environment features include the current polarization degree variance, the current polarization angle fluctuation amplitude, the current light intensity change rate, and the current turbidity value; The cosine similarity algorithm is used to calculate the global similarity between the current light field environment features and the standard scene features in the preset standard scene feature library; If the global similarity is greater than or equal to the preset similarity threshold of the target standard scene, then the target standard scene is determined as the current light field scene type.

4. The method according to claim 1, characterized in that, The step involves dynamically iteratively optimizing the baseline decision parameters corresponding to the current light field scene type based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, to obtain the optimal decision parameters, including: Based on the current light field scene type, determine the benchmark decision parameters corresponding to the current light field scene type; The particle swarm optimization algorithm is used to dynamically iterate and optimize the baseline decision parameters by combining the dynamic changes of the light field environment parameters in the target light field environment parameter set, so as to obtain the optimal decision parameters that are suitable for the current light field scene type.

5. The method according to claim 1, characterized in that, The step of verifying the navigation data and outputting a navigation decision instruction that passes the verification includes: The navigation data is fused and verified with the inertial navigation data and underwater acoustic navigation auxiliary data corresponding to the target underwater device to obtain the current polarization navigation error; If the current polarization navigation error is less than or equal to a preset navigation accuracy threshold, then the navigation data verification is determined to be successful, and the navigation decision instruction is generated based on the verified navigation data. If the current polarization navigation error is greater than the preset navigation accuracy threshold, the navigation data is discarded, and the process returns to the step of determining the optimal solution algorithm based on the current light field scene type, and combining the optimal decision parameters to solve the polarization parameters in the target light field environment parameter set to obtain navigation data, until the navigation decision command that has passed verification is output.

6. The method according to claim 1, characterized in that, The method further includes: During the process of light field scene recognition, the light field environment parameters continuously collected from the target light field environment parameter set are linearly fitted to obtain parameter change trend data. Based on the parameter change trend data and the preset parameter change threshold, light field anomalies are identified. If it is determined that there is an anomaly in the light field environment parameters in the target light field environment parameter set, a light field anomaly warning signal for the current light field scene type is output. After receiving the light field anomaly warning signal, the baseline decision parameters corresponding to the current light field scene type are automatically adjusted according to the preset anomaly adjustment strategy; Based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, the adjusted baseline decision parameters are dynamically iteratively optimized to obtain the optimal decision parameters.

7. An adaptive polarization navigation decision-making device, characterized in that, The device includes the following modules: The data acquisition and preprocessing module is used to acquire the original light field environment parameters of the underwater environment in which the target underwater device is located at the current moment, and to preprocess the original light field environment parameters to obtain the target light field environment parameter set. The light field scene recognition module is used to perform light field scene recognition based on the target light field environment parameter set to obtain the current light field scene type. The decision parameter optimization module is used to dynamically iteratively optimize the baseline decision parameters corresponding to the current light field scene type based on the current light field scene type and the dynamic changes of the light field environment parameters in the target light field environment parameter set, so as to obtain the optimal decision parameters. The polarization parameter calculation module is used to determine the optimal calculation algorithm based on the current light field scene type, and combine the optimal decision parameters to calculate the polarization parameters in the target light field environment parameter set to obtain navigation data. The navigation data verification module is used to verify the navigation data and output navigation decision instructions that have passed the verification.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.

Images