Real-time communication and positioning data fusion processing method for AUV (Autonomous Underwater Vehicle) collaborative operation

Abstract

The invention relates to the technical field of AUVs (Autonomous Underwater Vehicles), in particular to a real-time communication and positioning data fusion processing method for AUV (Autonomous Underwater Vehicle) collaborative operation, which comprises the following steps of: deploying a multi-modal sensor, acquiring communication data, kinematics data, acoustic positioning data and visual data required by an AUV, and preprocessing the data; the method comprises the following steps: fusing multi-modal data by using an extended Kalman filtering algorithm, calculating three-dimensional position and attitude information of the AUV in real time, carrying out dynamic path planning by using an ant colony optimization algorithm based on fused positioning data, optimizing a collaborative operation path between the AUVs, and sharing the positioning data and a path planning result in real time through an underwater communication network. And dynamically adjusting a task allocation strategy by using an improved task allocation algorithm, optimizing the cooperation efficiency, and finally, storing all data, path planning records and task allocation schemes in a database. According to the invention, the accuracy and stability of AUV collaborative operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous underwater robots, and in particular to a method for fusion processing of real-time communication and positioning data of AUV collaborative operation. Background Art

[0002] With the continuous development of autonomous underwater robot (AUV) technology, AUVs are increasingly used in underwater exploration, environmental monitoring, resource survey and other fields. AUVs can perform high-precision and high-efficiency tasks in complex underwater environments. Therefore, their collaborative operation capabilities have become the key to improving operational efficiency and accuracy. When multiple AUVs work together, they need to rely on efficient real-time data acquisition and processing, path planning and task allocation mechanisms. To achieve this goal, AUVs are usually equipped with multimodal sensors, covering acoustic communication, acoustic positioning, inertial navigation, Doppler velocity measurement, and visual camera functions.

[0003] However, although AUV collaborative operations have shown great potential in many fields, existing technologies still face many challenges, especially in data fusion, communication quality, path planning and task allocation. There are significant technical problems. First, existing multimodal data fusion methods usually rely on traditional sensor fusion algorithms, which are difficult to deal with the noise and missing problems of sensor data in dynamic underwater environments, resulting in low positioning accuracy. Secondly, the stability of underwater communication networks is poor, and communication channels are affected by water flow, noise and environmental factors. Existing communication optimization technologies have not been able to fully solve the problems of channel instability and delay, affecting the effects of real-time data sharing and path planning. In addition, existing path planning algorithms lack sufficient adaptability in the face of dynamic environmental changes, and path conflicts and low task coverage are common problems. Summary of the invention

[0004] The present invention provides a method for fusion processing of real-time communication and positioning data of AUV collaborative operation.

[0005] A method for fusion processing of real-time communication and positioning data of AUV collaborative operation, comprising the following steps: S1, real-time data acquisition and preprocessing: through the multimodal sensors deployed on the AUV, including acoustic communication unit, acoustic positioning unit, inertial navigation unit, Doppler velocimeter and camera equipment, the multimodal data required for collaborative operation are collected. The multimodal data includes communication data, kinematic data, acoustic positioning data and visual data. The collected multimodal data are preprocessed. The preprocessing includes denoising, interpolation and time series alignment to generate a multimodal data set; S2, dynamic communication quality assessment and optimization: Based on the collected communication data, channel information is extracted, and the stability and performance indicators of the communication channel are dynamically evaluated using an adaptive weight allocation algorithm, and the communication strategy is dynamically optimized in combination with channel conditions; S3, multi-modal positioning data fusion: Based on the collected kinematic data, acoustic positioning data and visual data, a dynamic factor graph model is constructed, and the kinematic data, acoustic positioning data and visual data are fused using the extended Kalman filter algorithm to calculate the three-dimensional position and attitude information of the AUV in real time to generate positioning data; S4, collaborative operation path planning and adjustment: Based on the positioning data generated in S3 and combined with the relative position relationship between AUVs, a collaborative path planning algorithm based on ant colony optimization is used to dynamically adjust the AUV operation path to avoid path conflicts; S5, real-time data sharing and task allocation: through the underwater communication network, the generated positioning data and path planning results are shared in real time to all AUV collaborative units. Using the improved task allocation algorithm, the task allocation strategy is dynamically adjusted according to the AUV operation status, path planning results and task priority to optimize the collaboration efficiency; S6, data storage: storing the collected and processed multimodal data, positioning data, path planning records and task allocation plans in the database.

[0006] Optionally, the S1 includes: S11, communication data collection: using the acoustic communication unit on the AUV, real-time communication channel data is collected, including signal-to-noise ratio, bandwidth utilization, delay and packet loss rate. The collected data is marked with a timestamp and recorded in real time; S12, kinematic data acquisition: The acceleration, angular velocity and heading angle of the AUV are collected by the inertial navigation unit, and the instantaneous linear velocity of the AUV is calculated by combining the velocity components collected by the Doppler velocimeter; S13, acoustic positioning data collection: The distance data and azimuth angle data between the AUV and surrounding reference points are collected by the acoustic positioning unit using a multi-point ranging method, and the position of the AUV is calculated by combining the ranging accuracy weight; S14, visual data acquisition: using video equipment to collect high-resolution images of the seabed environment, identifying key markers through feature point extraction algorithms, obtaining visual feature points, using optical flow methods to calculate the relative motion of visual feature points, and generating visual positioning information; S15, data preprocessing: preprocessing the collected multimodal data.

[0007] S16, Multimodal dataset generation: Integrate communication data, kinematic data, acoustic localization data, and visual data to generate a multimodal dataset.

[0008] Optionally, S2 includes: S21, channel information extraction: extract channel information from communication data for channel evaluation, the channel information includes signal-to-noise ratio, bandwidth utilization, delay and packet loss rate; S22, communication channel performance evaluation: Adopting adaptive weight allocation algorithm, dynamically calculate channel performance indicators according to the importance of communication tasks and environmental conditions; S23, adaptive weight adjustment: using gradient descent method to optimize channel performance indicators; S24, communication strategy optimization: dynamically adjust the communication strategy based on the optimized channel performance indicators.

[0009] Optionally, the S3 includes: S31, dynamic factor graph model construction: Based on kinematic data, acoustic positioning data and visual data, a dynamic factor graph model is constructed to describe the dependency between the state and observation of the AUV, which is expressed as: ; Among them, V includes state variable nodes and Observed Variable Node ; S32, state prediction: predict the state variable nodes according to the state transfer equation; S33, observation update: Combined with the observation variable node, the state variable node is updated through the observation equation; S34, multimodal data fusion: Fusion of kinematic data (high-frequency dynamic information), acoustic localization data (low-frequency absolute position information), and visual data (visual feature points) to generate state estimates; S35, generating positioning data: generating positioning data of the AUV based on the fused state estimation.

[0010] Optionally, the S4 includes: S41, modeling the relative position relationship between AUVs: determining the relative position relationship of each AUV, including relative distance, angle and relative motion state, based on the positioning data of the generated AUVs; S42, task area division: dividing the task area according to the relative positions between the AUVs and the size of the operation area; S43, ant colony optimization path planning initialization: based on the relative position of the AUV and the division of the task area, the ant colony optimization algorithm is used to initialize the path planning; S44, path adjustment and optimization: adjust the AUV’s path according to real-time data and environmental changes, optimize the operation path, and avoid path conflicts.

[0011] Optionally, S4 further includes: S45, real-time environmental monitoring: The environmental data of the operation area are monitored in real time through the environmental sensors on the AUV. The environmental data include water flow rate, obstacle location and water temperature, and an environmental model is constructed for path planning; S46, environmental model establishment: based on the real-time collected environmental data, a three-dimensional environmental model of the AUV operating environment is established to describe the obstacle distribution, water flow velocity field and temperature distribution in the operating area; S47, environmental model and path planning input: combining the established environmental model with the positioning data generated in S3 and the relative position relationship between the AUVs to form a comprehensive environmental model as input for path planning; S48, path planning optimization: Based on the comprehensive environment model, the path is dynamically adjusted through the ant colony optimization algorithm to avoid obstacle interference.

[0012] Optionally, the S5 includes: S51, Data sharing mechanism design: Design a data sharing mechanism through an underwater communication network between AUVs; S52, Communication protocol design: Design an adaptive communication protocol to ensure data sharing stability between different AUVs; S53, data synchronization and timing alignment: synchronizing and timing aligning the real-time collected positioning data, path planning data and task data; S54, data sharing feedback mechanism: After receiving the shared data, each AUV will verify the correctness of the data through the feedback mechanism and return the feedback result to the data source AUV.

[0013] Optionally, S5 further includes: S55, Task allocation strategy design: Design an improved task allocation algorithm to dynamically adjust task allocation based on the AUV’s operating status, path planning results, and task priority; S56, operation status evaluation: the operation status of each AUV is evaluated in real time, including battery power, task completion progress and path execution status. The evaluation results are used to dynamically adjust task allocation. For example, if an AUV has low battery or slow task completion progress, it will receive more low-priority tasks. S57, Dynamic adjustment of task priority: Dynamically adjust the priority of tasks according to their actual urgency, resource requirements and environmental changes; S58, task allocation and adjustment feedback: According to the real-time operation status and path planning results of the AUV, task allocation is dynamically adjusted through the feedback mechanism.

[0014] Optionally, the S6 includes: S61, data classification and structured storage: classify the collected and processed multimodal data, positioning data, path planning records and task allocation plans, and use different storage structures according to different data types; S62, Database table design and index optimization: Design database table structure to support storage and retrieval of different data types; S63, data compression: compress the stored data, especially for large-scale visual data, path planning records and other unstructured data; S64, Data access control: Set up a permission access control mechanism to ensure that only authorized personnel can access, modify or delete data in the database, and use encryption technology to protect the privacy of stored data to prevent data leakage or tampering.

[0015] Beneficial effects of the present invention: The present invention, combined with multimodal sensors (including acoustic communication units, inertial navigation units, acoustic positioning units, etc.), effectively collects the communication data, kinematic data, acoustic positioning data and visual data required by AUV. Through pre-processing measures such as denoising, interpolation and time alignment, problems such as inconsistent data quality, missing data and noise interference can be solved. In particular, on the basis of real-time processing and multimodal data fusion, the high quality and reliability of data are ensured, and high-precision input data is provided for subsequent positioning data fusion, path planning and task allocation, thereby greatly improving the accuracy and stability of AUV collaborative operations.

[0016] The present invention, based on the dynamic communication quality evaluation and optimization method of the adaptive weight allocation algorithm, can dynamically optimize the communication strategy according to the stability and performance indicators (such as signal-to-noise ratio, delay and packet loss rate) of the communication channel. This real-time communication optimization not only improves the efficiency and reliability of data transmission, but also ensures the efficient sharing of data between AUVs, providing a guarantee for the real-time update of path planning and task allocation. In addition, by introducing an algorithm based on ant colony optimization, combined with the real-time positioning data and task requirements of AUVs, the path can be dynamically adjusted to avoid path conflicts and improve task coverage. Combined with the real-time data sharing and task allocation strategy in S5, the collaboration efficiency between AUVs can be optimized, ensuring the reasonable allocation of task priorities, thereby improving the overall work efficiency and collaborative work capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of a method flow of an embodiment of the present invention; Figure 2 Schematic diagram of S4 process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0020] like Figure 1-Figure 2 As shown, a method for fusion processing of real-time communication and positioning data of AUV collaborative operation includes the following steps: S1, real-time data acquisition and preprocessing: through the multimodal sensors deployed on the AUV, including acoustic communication unit, acoustic positioning unit, inertial navigation unit, Doppler velocimeter and camera equipment, the multimodal data required for collaborative operation are collected. The multimodal data includes communication data, kinematic data, acoustic positioning data and visual data. The collected multimodal data are preprocessed. The preprocessing includes denoising, interpolation and time series alignment to generate a multimodal data set; S2, dynamic communication quality assessment and optimization: Based on the collected communication data, channel information is extracted, and the stability and performance indicators of the communication channel are dynamically evaluated using an adaptive weight allocation algorithm, and the communication strategy is dynamically optimized in combination with channel conditions; S3, multi-modal positioning data fusion: Based on the collected kinematic data, acoustic positioning data and visual data, a dynamic factor graph model is constructed, and the kinematic data, acoustic positioning data and visual data are fused using the extended Kalman filter algorithm to calculate the three-dimensional position and attitude information of the AUV in real time and generate positioning data. The fusion process is guaranteed by the communication strategy optimized in S2 to ensure the integrity and real-time performance of the transmitted data, effectively reduce data loss and positioning errors caused by unstable communication, and provide accurate input for path planning; S4, collaborative operation path planning and adjustment: Based on the positioning data generated in S3 and combined with the relative position relationship between AUVs, a collaborative path planning algorithm based on ant colony optimization is used to dynamically adjust the AUV operation path to avoid path conflicts and improve task coverage; S5, real-time data sharing and task allocation: through the underwater communication network, the generated positioning data and path planning results are shared in real time to all AUV collaborative units. Using the improved task allocation algorithm, the task allocation strategy is dynamically adjusted according to the AUV operation status, path planning results and task priority to optimize the collaboration efficiency; S6, data storage: storing the collected and processed multimodal data, positioning data, path planning records and task allocation plans in the database.

[0021] S1 includes: S11, communication data collection: using the acoustic communication unit on the AUV, real-time communication channel data is collected, including signal-to-noise ratio, bandwidth utilization, delay and packet loss rate. The collected data is marked with a timestamp and recorded in real time; S12, Kinematic data acquisition: Acceleration of the AUV is collected through the inertial navigation unit , angular velocity and heading angle , combined with the velocity component collected by the Doppler velocimeter , , , calculate the instantaneous linear velocity of the AUV , the calculation formula is expressed as: ; The motion trajectory of the AUV is calculated through three-axis acceleration to provide high-frequency dynamic support for positioning data; S13, Acoustic Positioning Data Collection: The distance data between the AUV and surrounding reference points is collected by using the acoustic positioning unit and the multi-point ranging method. and direction angle data , and combined with the ranging accuracy weight , calculate the position of the AUV , expressed as: ; in, represents the three-dimensional coordinates of the reference point, Proportional to the ranging accuracy; S14, Visual Data Collection: Use camera equipment to collect high-resolution images of the seabed environment, identify key landmarks through feature point extraction algorithms, and obtain visual feature points ,The optical flow method is used to calculate the relative motion of the visual feature points and generate the visual positioning information; S15, data preprocessing: preprocessing the collected multimodal data, including the following specific steps: Denoising: Wavelet transform is used to filter out high-frequency noise in multimodal data to ensure data smoothness; Interpolation processing: For data points with large sampling gaps, cubic spline interpolation algorithm is used to complete the data; Time series alignment: Unify the time axis of multimodal data based on timestamps to ensure that all data have a corresponding relationship at the same time and generate a standardized multimodal data set.

[0022] S16, multimodal dataset generation: integrating communication data, kinematic data, acoustic localization data, and visual data to generate a multimodal dataset; Through real-time data collection and preprocessing, the problems of noise interference, incomplete data and time asynchrony in the multimodal data collection process are solved, and a high-quality multimodal data set is generated, providing a solid data foundation for subsequent steps of communication optimization, data fusion and path planning.

[0023] S2 includes: S21, channel information extraction: extract channel information from communication data for channel evaluation, the channel information includes signal-to-noise ratio, bandwidth utilization, delay and packet loss rate, where; Signal-to-noise ratio: using the collected communication signal power and noise power , calculate the signal-to-noise ratio, expressed as: ; Bandwidth utilization: The actual bandwidth used by the communication link With the maximum available bandwidth The ratio calculation is expressed as: ; Delay: measures the time difference between when a signal is sent and when it is received, expressed as: ; Packet loss rate: It is determined by calculating the ratio of the number of lost data packets to the total number of data packets, expressed as: ; S22, communication channel performance evaluation: Adopting the adaptive weight allocation algorithm, the channel performance index Q is dynamically calculated according to the importance of the communication task and the environmental conditions, which is expressed as: ; in, , , , is the weight factor, satisfying: ; The weight factor is dynamically adjusted according to the real-time requirements of the communication task, and the initial value is obtained through historical data training; S23, Adaptive weight adjustment: Use the gradient descent method to optimize the channel performance index, specifically, use the gradient descent method to optimize the weight factor , , , , the goal is to minimize the channel performance index Q and the target value The deviation is expressed as: ; The weight update formula is expressed as: ; in, is the learning rate; S24, communication strategy optimization: based on the optimized channel performance indicators, dynamically adjust the communication strategy, specifically including: Priority adjustment: According to the real-time requirements of the task, key data (such as task status and positioning data) are transmitted first, and the transmission frequency of low-priority data is reduced; Redundant coding strategy: Use low-density parity check coding algorithm to perform redundant processing on high-priority data to improve transmission reliability. The coding gain formula is: ; in, and are the data transmission rates of uncoded and coded data, respectively; Through dynamic communication quality evaluation and optimization, the performance status of the communication channel can be analyzed in real time, and the communication strategy can be dynamically adjusted to ensure the stability of the communication link and the reliability of data transmission in multi-AUV collaborative operations, providing strong communication support for data fusion and path planning.

[0024] S3 includes: S31, dynamic factor graph model construction: Based on kinematic data, acoustic positioning data and visual data, a dynamic factor graph model is constructed to describe the dependency between the state and observation of the AUV, which is expressed as: ; Among them, V includes state variable nodes and Observed Variable Node , state variable node , represents the position, attitude and other state information of the AUV at k, for example: ; in, , , represents the position coordinates of the AUV in three-dimensional space, , is the horizontal coordinate, is the vertical coordinate (depth), is the pitch angle, which indicates the tilt angle of the AUV in the longitudinal direction. is the roll angle, which indicates the roll angle of the AUV in the lateral direction. is the yaw angle, which indicates the rotation angle of the AUV on the horizontal plane; Observation Variable Node , represents the observation data of AUV at time k, for example: ; in, Indicates the acceleration value collected by the inertial navigation unit, Represents the distance information measured by the acoustic positioning unit. Represents the extracted visual feature points; E represents factor node, including state transfer factor and observation factor , state transfer factor Describes the time change of the state, the observation factor Describe the relationship between states and observations; S32, state prediction: predict the state variable node according to the state transfer equation, expressed as: ; in, is the state transfer equation, describing arrive changes, is the process noise, satisfying ; S33, observation update: Combined with the observation variable node, the state variable node is updated through the observation equation, which is expressed as: (1) Calculate the observation prediction, expressed as: ; (2) Calculate the Kalman gain, expressed as: ; in, is the Jacobian matrix of the observation equation; (3) Update the state variables, expressed as: ; in, is the observation equation, describing and , R is the observation noise covariance matrix; S34, multimodal data fusion: Fusion of kinematic data (high-frequency dynamic information), acoustic localization data (low-frequency absolute position information), and visual data (visual feature points) to generate state estimates; The fusion process controls the contribution of each data source to the final state by weight adjustment, and the weight is dynamically calculated based on the observed noise covariance R. The final state expression is: ; in, is the observation data from the i-th data source, is the weight factor, which is dynamically adjusted according to the sensor accuracy; S35, generating positioning data: generating positioning data of the AUV according to the fused state estimation, including; Location data : represents the coordinates of AUV in three-dimensional space; Attitude data (pitch angle , Roll Angle , yaw angle ): indicates the spatial direction of AUV; Positioning data is marked with a timestamp to ensure the accuracy of subsequent route planning and real-time sharing; Combined with the optimized communication strategy in S2, it ensures that kinematic data, acoustic positioning data and visual data are transmitted without packet loss and with low latency, thus ensuring the real-time and integrity of the fusion process.

[0025] S4 includes: S41, modeling the relative position relationship between AUVs: determining the relative position relationship of each AUV, including relative distance, angle and relative motion state, based on the positioning data of the generated AUVs; The relative distance is expressed as: ; in, is the relative distance between AUVi and AUVj, , , is the position coordinate of AUVi, , , is the position coordinate of AUVj; The relative angle is expressed as: ; in, is the relative angle between AUVi and AUVj; S42, task area division: according to the relative positions between AUVs and the size of the operation area, the task area is divided to ensure that the operation area of ​​each AUV does not overlap and the task coverage is maximized. Each AUV performs path planning in its task area. The calculation formula of the task area is: ; in, is the mission area of ​​AUVi, is the radius that AUVi can cover; S43, ant colony optimization path planning initialization: Based on the relative position of the AUV and the division of the task area, the ant colony optimization algorithm is used to initialize the path planning. The path selection probability formula is expressed as: ; in, is the path pheromone concentration, is the path heuristic information, and To adjust the parameters, is the set of adjacent nodes of AUVi; S44, path adjustment and optimization: According to real-time data and environmental changes, adjust the path of the AUV, optimize the operation path, avoid path conflicts, and improve task coverage. The path optimization formula is expressed as: ; in, is the distance between node i and node i+1, is the risk value of the path segment, is the risk weight factor.

[0026] S4 also includes: S45, real-time environmental monitoring: The environmental data of the operation area is monitored in real time through the environmental sensors on the AUV (such as water flow sensor, temperature sensor, obstacle detection sensor, etc.). The environmental data includes water flow rate, obstacle location and water temperature, and an environmental model is constructed for path planning; S46, environmental model establishment: based on the real-time collected environmental data, a three-dimensional environmental model of the AUV operating environment is established to describe the obstacle distribution, water flow velocity field and temperature distribution in the operating area. The environmental model can be represented by gridding, and each grid node contains data on the existence of obstacles and other environmental factors; S47, environmental model and path planning input: the established environmental model is combined with the positioning data generated in S3 and the relative position relationship between the AUVs to form a comprehensive environmental model as the input of the path planning. The path planning takes into account the navigation constraints of the AUV, the influence of obstacles and the interference of water flow; S48, path planning optimization: Based on the comprehensive environment model, the path is dynamically adjusted through the ant colony optimization algorithm to avoid obstacle interference and ensure the shortest path and task coverage. The optimization formula is expressed as: ; in, is the distance of the initial path, For the optimized path distance, the goal is to minimize the path length.

[0027] S5 includes: S51, Data Sharing Mechanism Design: Design a data sharing mechanism through the underwater communication network between AUVs to ensure that positioning data, path planning results and task execution status can be shared in real time. The data transmission adopts an adaptive adjustment strategy to automatically adjust the transmission frequency, transmission rate and redundancy when the channel conditions change. The basic structure of data sharing includes: ; in, A shared data set, including: Indicates real-time positioning data, Represents the path planning results and Indicates the task execution status; S52, Communication protocol design: Design an adaptive communication protocol to ensure the stability of data sharing between different AUVs, especially when the environment is complex or the communication channel is unstable. The communication protocol includes mechanisms such as data packaging, encoding and decoding, and error checking. The transmission protocol can adopt a distributed transmission strategy and support dynamic bandwidth allocation; S53, data synchronization and timing alignment: synchronize and timing-align the real-time collected positioning data, path planning data and task data to ensure that all data are synchronized at the same time point to avoid data conflicts caused by delays. The data synchronization process includes timestamp marking and interpolation processing; S54, data sharing feedback mechanism: After receiving the shared data, each AUV will verify the correctness of the data through the feedback mechanism and return the feedback result to the data source AUV. The feedback information includes data integrity check, delay feedback and communication quality report.

[0028] The S5 also includes: S55, Task allocation strategy design: Design an improved task allocation algorithm, dynamically adjust task allocation based on the AUV's operating status, path planning results, and task priority. The goal of task allocation is to maximize task coverage and improve the AUV's collaborative efficiency. The task allocation process uses the following objective function: ; in, is the execution time of task i, is the priority of task i, and the goal is to minimize the total execution time of the task; S56, operation status evaluation: the operation status of each AUV is evaluated in real time, including battery power, task completion progress and path execution status. The evaluation results are used to dynamically adjust task allocation. For example, if an AUV has low battery or slow task completion progress, it will receive more low-priority tasks. S57, Dynamic adjustment of task priority: Dynamically adjust the task priority according to the actual urgency of the task, resource requirements and environmental changes. The task priority evaluation formula is: ; in, is the priority of the task, The urgency of the task, is the deadline of the task, For task execution efficiency, , , is the weight coefficient; S58, task allocation and adjustment feedback: According to the real-time operation status and path planning results of the AUV, task allocation is dynamically adjusted through the feedback mechanism. The feedback includes the progress report of task execution, path execution status and task conflict detection. Based on the feedback information, the task allocation algorithm is adjusted to ensure efficient execution of the task.

[0029] S6 includes: S61, data classification and structured storage: Classify the collected and processed multimodal data, positioning data, path planning records and task allocation plans, and use different storage structures according to different data types. Specifically, all data will be classified according to timestamp, data type and AUV identification to generate the following four types of storage objects: Multimodal data: including communication data, kinematic data, acoustic positioning data and visual data.

[0030] Positioning data: includes the three-dimensional position and attitude information of each AUV.

[0031] Path planning records: including path planning results, optimization information and path adjustment records when the AUV performs tasks.

[0032] Task allocation plan: including detailed information on task allocation, such as the assigned AUV for each task, task status, and priority; S62, Database table design and index optimization: Design database table structure to support storage and retrieval of different data types; S63, data compression: compress the stored data, especially for large-scale visual data, path planning records and other unstructured data, using appropriate compression algorithms (such as JPEG, PNG compression of image data, or Huffman coding to compress path planning record data); S64, Data access control: Set up a permission access control mechanism to ensure that only authorized personnel can access, modify or delete data in the database, and use encryption technology to protect the privacy of stored data to prevent data leakage or tampering.

[0033] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0034] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for fusion processing of real-time communication and positioning data for AUV collaborative operation, characterized in that: The following steps are involved: S1, real-time data acquisition and preprocessing: through the multimodal sensors deployed on the AUV, including acoustic communication unit, acoustic positioning unit, inertial navigation unit, Doppler velocimeter and camera equipment, the multimodal data required for collaborative operation are collected. The multimodal data includes communication data, kinematic data, acoustic positioning data and visual data. The collected multimodal data are preprocessed. The preprocessing includes denoising, interpolation and time series alignment to generate a multimodal data set; S2, dynamic communication quality assessment and optimization: Based on the collected communication data, channel information is extracted, and the stability and performance indicators of the communication channel are dynamically evaluated using an adaptive weight allocation algorithm, and the communication strategy is dynamically optimized in combination with channel conditions; S3, multi-modal positioning data fusion: Based on the collected kinematic data, acoustic positioning data and visual data, a dynamic factor graph model is constructed, and the kinematic data, acoustic positioning data and visual data are fused using the extended Kalman filter algorithm to calculate the three-dimensional position and attitude information of the AUV in real time to generate positioning data; S4, collaborative operation path planning and adjustment: Based on the positioning data generated in S3 and combined with the relative position relationship between AUVs, a collaborative path planning algorithm based on ant colony optimization is used to dynamically adjust the AUV operation path to avoid path conflicts; S5, real-time data sharing and task allocation: through the underwater communication network, the generated positioning data and path planning results are shared in real time to all AUV collaborative units. Using the improved task allocation algorithm, the task allocation strategy is dynamically adjusted according to the AUV operation status, path planning results and task priority to optimize the collaboration efficiency; S6, data storage: storing the collected and processed multimodal data, positioning data, path planning records and task allocation plans in the database.

2. The method for fusion processing of real-time communication and positioning data for AUV collaborative operation according to claim 1 is characterized in that: The S1 includes: S11, communication data collection: using the acoustic communication unit on the AUV, real-time communication channel data is collected, including signal-to-noise ratio, bandwidth utilization, delay and packet loss rate. The collected data is marked with a timestamp and recorded in real time; S12, kinematic data acquisition: The acceleration, angular velocity and heading angle of the AUV are collected by the inertial navigation unit, and the instantaneous linear velocity of the AUV is calculated by combining the velocity components collected by the Doppler velocimeter; S13, acoustic positioning data collection: The distance data and azimuth angle data between the AUV and surrounding reference points are collected by the acoustic positioning unit using a multi-point ranging method, and the position of the AUV is calculated by combining the ranging accuracy weight; S14, visual data acquisition: using a camera to collect images of the seabed environment, identifying landmarks through a feature point extraction algorithm, obtaining visual feature points, using an optical flow method to calculate the relative motion of the visual feature points, and generating visual positioning information; S15, data preprocessing: preprocessing the collected multimodal data; S16, Multimodal dataset generation: Integrate communication data, kinematic data, acoustic localization data, and visual data to generate a multimodal dataset.

3. The method for fusion processing of real-time communication and positioning data of AUV collaborative operation according to claim 2 is characterized in that: The S2 includes: S21, channel information extraction: extract channel information from communication data for channel evaluation, the channel information includes signal-to-noise ratio, bandwidth utilization, delay and packet loss rate; S22, communication channel performance evaluation: Adopting adaptive weight allocation algorithm, dynamically calculate channel performance indicators according to the importance of communication tasks and environmental conditions; S23, adaptive weight adjustment: using gradient descent method to optimize channel performance indicators; S24, communication strategy optimization: dynamically adjust the communication strategy based on the optimized channel performance indicators.

4. The method for fusion processing of real-time communication and positioning data for AUV collaborative operation according to claim 3 is characterized in that: The S3 includes: S31, dynamic factor graph model construction: Based on kinematic data, acoustic positioning data and visual data, a dynamic factor graph model is constructed to describe the dependency between the state and observation of the AUV, which is expressed as: ； Among them, V includes state variable nodes and Observed Variable Node ; S32, state prediction: predict the state variable nodes according to the state transfer equation; S33, observation update: Combined with the observation variable node, the state variable node is updated through the observation equation; S34, multimodal data fusion: fusing kinematic data, acoustic localization data, and visual data to generate state estimates; S35, generating positioning data: generating positioning data of the AUV based on the fused state estimation.

5. The method for fusion processing of real-time communication and positioning data of AUV collaborative operation according to claim 4 is characterized in that: The S4 includes: S41, modeling the relative position relationship between AUVs: determining the relative position relationship of each AUV, including relative distance, angle and relative motion state, based on the positioning data of the generated AUVs; S42, task area division: dividing the task area according to the relative positions between the AUVs and the size of the operation area; S43, ant colony optimization path planning initialization: based on the relative position of the AUV and the division of the task area, the ant colony optimization algorithm is used to initialize the path planning; S44, path adjustment and optimization: adjust the AUV’s path according to real-time data and environmental changes, optimize the operation path, and avoid path conflicts.

6. The method for fusion processing of real-time communication and positioning data of AUV collaborative operation according to claim 5 is characterized in that: The S4 further comprises: S45, real-time environmental monitoring: The environmental data of the operation area are monitored in real time through the environmental sensors on the AUV. The environmental data include water flow rate, obstacle location and water temperature, and an environmental model is constructed for path planning; S46, environmental model establishment: based on the real-time collected environmental data, a three-dimensional environmental model of the AUV operating environment is established to describe the obstacle distribution, water flow velocity field and temperature distribution in the operating area; S47, environmental model and path planning input: combining the established environmental model with the positioning data generated in S3 and the relative position relationship between the AUVs to form a comprehensive environmental model as input for path planning; S48, path planning optimization: Based on the comprehensive environment model, the path is dynamically adjusted through the ant colony optimization algorithm to avoid obstacle interference.

7. The method for fusion processing of real-time communication and positioning data of AUV collaborative operation according to claim 6 is characterized in that: The S5 includes: S51, Data sharing mechanism design: Design a data sharing mechanism through an underwater communication network between AUVs; S52, Communication protocol design: Design an adaptive communication protocol to ensure data sharing stability between different AUVs; S53, data synchronization and timing alignment: synchronizing and timing aligning the real-time collected positioning data, path planning data and task data; S54, data sharing feedback mechanism: After receiving the shared data, each AUV will verify the correctness of the data through the feedback mechanism and return the feedback result to the data source AUV.

8. The method for fusion processing of real-time communication and positioning data for AUV collaborative operation according to claim 7, characterized in that: The S5 further includes: S55, Task allocation strategy design: Design an improved task allocation algorithm to dynamically adjust task allocation based on the AUV’s operating status, path planning results, and task priority; S56, operation status evaluation: real-time evaluation of the operation status of each AUV, including battery power, task completion progress and path execution status. The evaluation results are used to dynamically adjust task allocation; S57, Dynamic adjustment of task priority: Dynamically adjust the priority of tasks according to their actual urgency, resource requirements and environmental changes; S58, task allocation and adjustment feedback: According to the real-time operation status and path planning results of the AUV, task allocation is dynamically adjusted through the feedback mechanism.

9. The method for fusion processing of real-time communication and positioning data for AUV collaborative operation according to claim 8, characterized in that: The S6 includes: S61, data classification and structured storage: classify the collected and processed multimodal data, positioning data, path planning records and task allocation plans, and use different storage structures according to different data types; S62, Database table design and index optimization: Design database table structure to support storage and retrieval of different data types; S63, data compression: compressing the stored data; S64, Data access control: Set up permission access control mechanism and use encryption technology to protect the privacy of stored data.

Images