Distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking

Abstract

The invention discloses a distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking, which is a control unit which is carried in a UUV, is powered by the UUV and communicates with the UUV and a shore-based platform, and software covers a communication sub-module, a mission analysis sub-module, a perception information processing sub-module, a behavior decision sub-module, a cooperative algorithm sub-module, a fault detection sub-module, a data recording sub-module and the like. The communication sub-module provides a standard interface protocol to realize information exchange and isomerization support; the mission analysis sub-module analyzes a static or dynamic task instruction; the sensing information processing sub-module processes sonar data and transmits the sonar data to the behavior decision sub-module for behavior planning, the collaborative algorithm sub-module outputs an instruction, the fault detection sub-module judges a fault and gives a processing instruction, and the data recording sub-module stores key data. Based on a sonar design processing and decision-making method and packaging into a module, heterogeneous UUV cooperative task efficiency is improved, target detection and tracking tasks are completed, and multi-UUV cooperative operation capability is enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of cooperative control of multiple underwater unmanned vehicles, and relates to a distributed heterogeneous cooperative control module of multiple underwater unmanned vehicles for target detection and tracking. Background Art

[0002] In recent years, with the rapid development of underwater unmanned vehicle (UUV) technology, UUVs are increasingly used in marine exploration, environmental monitoring, underwater rescue, resource surveys, and military fields. In complex underwater environments, the capabilities and coverage of a single UUV are usually limited, while multiple heterogeneous UUVs can significantly improve the efficiency and quality of tasks through collaborative work. Therefore, how to achieve efficient collaboration of heterogeneous UUVs in collaborative tasks has become a hot topic and difficulty in current research.

[0003] In the collaborative detection and target tracking tasks of heterogeneous UUVs, it is usually necessary to accurately locate and detect static targets such as shipwrecks and mineral resource points, and dynamically respond to the motion trajectory of underwater mobile targets such as submersibles. Such tasks place high demands on the perception, decision-making and collaborative control of UUVs. However, due to the differences in hardware, interface protocols and instruction sets between different UUVs, traditional control methods often cannot meet the needs of multi-UUV collaborative tasks. For example, the low degree of standardization of communication protocols between UUVs makes it difficult to transmit collaborative instructions; different processing methods of perception data affect the efficiency of information sharing and comprehensive decision-making; the lack of a unified framework for behavioral decision-making and control strategies restricts the flexibility and scalability of tasks.

[0004] In response to these challenges, academia and industry have proposed some methods, such as global optimization methods based on centralized control, autonomous decision-making strategies based on distributed control, and multi-sensor fusion technology. However, most of these methods rely on specific hardware architectures or control systems, lack versatility, and are difficult to adapt to the collaborative operation requirements of multiple heterogeneous UUVs. At the same time, the real-time and robustness of existing methods in complex underwater environments still need to be improved, especially in static target detection and dynamic target tracking tasks, the coordination and cooperation between UUVs has not been effectively solved.

[0005] In order to solve the above problems, the present invention proposes a collaborative control module for target detection and tracking tasks. The module is based on a unified input and output interface protocol and a standardized instruction set, and is compatible with various types of UUVs. Its core design is based on the perception data obtained by the sonar detection sensor, combined with advanced behavioral decision logic and collaborative algorithms, to support collaborative operations between heterogeneous UUVs, and significantly improve the efficiency of the cluster in completing target detection and tracking tasks. At the same time, the present invention adopts a modular design, is highly compatible with the existing UUV hardware system, does not interfere with the underlying control logic of the UUV, fully retains the independence and stability of the existing system, and shows significant advantages in multi-UUV collaborative application scenarios, providing an innovative solution for multi-UUV collaborative detection and tracking. Summary of the invention

[0006] To solve the above technical problems, the purpose of the present invention is to provide a distributed heterogeneous multi-underwater unmanned vehicle collaborative control module for target detection and tracking, so as to realize the system design, development and application of multi-heterogeneous UUV collaborative detection of static targets and tracking of dynamic targets under unified standards.

[0007] The specific technical solutions are as follows: A distributed heterogeneous multi-underwater unmanned vehicle collaborative control module for target detection and tracking, the collaborative control module is installed in a UUV platform sealed cabin and powered by a UUV power management module, exchanges data with a UUV platform communication module and a shore-based platform through a unified standard communication protocol, and does not participate in the UUV bottom-level control. Multiple UUVs are each equipped with a collaborative control module to form a distributed multi-UUV collaborative control system, and the collaborative control module includes: The communication submodule provides a unified standard input and output interface for exchanging information with the UUV carrier and shore-based platform; Mission parsing submodule, which parses mission instructions based on a unified standard instruction set and supports static and dynamic mission modes; The perception information processing submodule receives and processes target detection data from the sonar sensor, including target angle, signal-to-noise ratio, and intensity; The behavior decision submodule determines the behavior strategy of the UUV based on the processed perception information and task instructions; The collaborative algorithm submodule executes the control algorithm and generates unified standard UUV control instructions; the fault detection and processing submodule monitors system errors and generates fault processing instructions; The data recording submodule stores key system data for other submodules to call and store to support subsequent analysis.

[0008] The communication submodule comprises: Input mission information: including the communication topology configuration of multiple UUVs, target operation area and target mission, and output to the mission analysis submodule; Input the node status information and the navigation status information of the module carrier UUV, including UTC timestamp, position, speed, attitude and other key information, which are used for collaborative control calculation; Input the sensor detection information of this node: the detection data received by the sonar detection sensor carried by the module carrier UUV, including key information such as target angle, signal-to-noise ratio and intensity, which is used for target state solution and tracking; Input neighbor node status information: The navigation status information of the neighboring UUV nodes, including UTC timestamp, position, speed, attitude and other key information, which is used for collaborative control calculation; Input the target perception information of neighboring nodes: The neighboring UUV nodes use their own perception information processing submodules to filter and process the sensor detection data and estimate the target information, including key information such as target speed, target heading, target relative angle and target relative distance, which are used for UUV behavior planning and control; Output the status information of this node: the input information of neighboring nodes, including the timestamp, position, speed, posture and other key information of this node, which is used for collaborative control calculation; Output the target perception information of this node: the input information of neighboring nodes, including the key information such as target detection angle, target solution position, target solution speed, target solution heading and confidence after optimization by the perception information processing submodule of this node, which is used for collaborative target detection and tracking; Output collaborative control command information: The control command information of the UUV calculated by collaborative control includes the expected speed, navigation and depth information that the UUV body needs to track. A unified standardized command set is used to realize the parsing and tracking of control commands of heterogeneous UUV platforms.

[0009] The mission analysis submodule includes: Static mission mode, used for area coverage search of static targets, the execution process has been fully pre-planned before the mission begins, that is, each stage of the mission and the execution order are fixed; The dynamic mission mode is used for dynamic target search and tracking tasks. It is dynamically determined based on real-time environmental data or changes in mission target status, and is adjusted in real time based on actual conditions.

[0010] The perception information processing submodule includes: receiving task instruction information from the collaborative algorithm submodule to determine the execution task mode, receiving target perception information of neighboring nodes and detection information of the node sensor from the communication submodule, fitting and filtering the detection angle using the sliding time window and least squares method, obtaining the detection angle and the confidence of the detection angle, estimating the target speed, heading, relative distance and the confidence of the state estimation based on the boundary constraint particle filter algorithm, selecting the perception information with high confidence as the trusted data after fusing the perception information of neighboring nodes, and passing the processed perception information to the behavior decision submodule for behavior planning and control.

[0011] The behavioral decision submodules include: The behavior decision submodule receives the perception information received from the perception information processing submodule, selects an appropriate behavior strategy according to the current state of the UUV, and determines whether the executed task is static or dynamic. When the executed task is a static task, behavior planning is performed according to a preset process, and at least one of the following behavior strategies is adopted: target search behavior decision, which is used to relocate the target when the target is lost; target tracking decision based on pure bearing, which is used to track when the resolution of target motion elements fails; target tracking decision based on motion behavior analysis, which is used to track when the resolution of target motion elements is valid.

[0012] The collaborative algorithm submodule, combined with the relevant historical key data information in the data recording submodule, processes the task instruction information output by the current mission analysis submodule, the node status information received from the UUV carrier in the communication submodule, the status information of each neighboring node of the underwater acoustic communication, the behavior control instructions output by the behavior decision submodule, etc., outputs unified standard control instructions to the communication submodule, and transmits them to the bottom control unit of the UUV carrier to realize UUV motion control.

[0013] The perception information processing submodule further converts the neighbor node perception information into data in the coordinate system of the current node when necessary.

[0014] The advantages of the present invention are: 1. Provide a universal unified standard interface protocol and distributed collaborative control system architecture, which can be carried on heterogeneous UUV platforms to realize distributed heterogeneous multi-UUV collaborative control and better complete collaborative detection and tracking tasks; 2. Provide a unified standard universal instruction set, so that each UUV can formulate mission tasks, analyze mission task instructions and control instructions under the unified standard, reducing the difficulty of system design, development, debugging and application; 3. Supports two task modes, static and dynamic, for target detection and tracking tasks. The static task supports preset process tasks such as multi-UUV collaborative formation and static target search in the collaborative area, while the dynamic task supports non-completely preset process tasks such as search, detection and tracking of dynamic targets in the collaborative area, which improves the flexibility and adaptability of multi-UUV collaborative tasks; 4. As an intermediate bridge module between upper-level decision-making and lower-level control, it is responsible for integrating distributed collaborative control systems, standardized input and output interface protocols, general instruction sets, and fault detection functions. Through data interaction with the communication module of the UUV body, this module realizes functions such as task parsing, instruction distribution, and collaborative control, while avoiding direct intervention in the underlying control logic, thereby effectively shielding the differences in types, interface openness, etc. of heterogeneous UUVs, simplifying the difficulty of system design and development, and improving the versatility, adaptability, and portability of the collaborative control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 : Schematic diagram of the installation and connection relationship of the collaborative control module inside the UUV carrier; Figure 2 : Overall flow chart of distributed heterogeneous multi-UUV collaborative task execution; Figure 3 : Task execution flow chart of regional search and tracking of dynamic targets; Figure 4 : Topological relationship diagram of each software sub-module of the collaborative control module; Figure 5 : Schematic diagram of searching for dynamic targets based on S-type maneuvers; Figure 6 : Schematic diagram of target tracking based on pure bearing and motion behavior analysis. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] The present invention provides a distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking, such as Figure 1As shown, each collaborative control module is independently packaged, with a total of three interfaces: communication port (including network port and serial port) and power supply port. It is installed inside the UUV carrier, powered by the UUV carrier power supply module, and connected to the communication module on the main control to realize network port or serial port communication. Each UUV constituting the cluster is equipped with the collaborative control module of the present invention, and the surface mother ship / shore-based command and control center uniformly issues mission tasks through WIFi or radio to collaboratively and efficiently complete the planned tasks.

[0018] Figure 2 It is a general flow chart of multi-UUV mission execution, which includes UUV deployment, preparation for navigation, navigation to the mission area, mission execution and the final return and wait for recovery stage. Among them, the dynamic and static tasks described in the present invention are mainly distinguished in the mission execution stage; static tasks include regional coverage search static target tasks, because their task execution process is only based on the preset execution of the planned tasks before preparation for navigation, and will not be dynamically adjusted according to the actual situation, so the present invention classifies them as static tasks. In this scenario, each UUV cooperates to conduct coverage search and detection of the area according to the planned mission tasks, and returns to the recovery point to wait for recovery after the coverage search is completed. Dynamic tasks include regional search and tracking dynamic target tasks, which are performed by Figure 3 It can be seen that since its mission execution process is not completely preset, but is adjusted in real time according to the mission target status, the present invention classifies it as a dynamic task. In this scenario, each UUV cooperates to detect the area according to the initial planned mission task. If a target is detected, it automatically switches to the tracking stage, and at the same time transmits the target information to the surrounding UUV neighbor nodes through underwater acoustic communication, and mobilizes the nearest neighbors to track collaboratively. In the tracking stage, the appropriate tracking mode is automatically switched based on the detection information of the sensor and the analysis information of the target motion behavior, including (1) if the detection information is valid but the target motion behavior analysis fails, then the pure bearing target tracking mode is selected; (2) if both the detection information and the target motion behavior analysis information are valid, then the tracking mode based on the target motion behavior analysis is selected; if the above-mentioned relevant information of the target is missing for a long time during the tracking stage, it will automatically switch back to the search stage to perform regional collaborative search.

[0019] Figure 4 The topological relationship of the software submodules of the collaborative control module is described. The collaborative control module software part includes the communication submodule, mission analysis submodule, perception information processing submodule, behavior decision submodule, collaborative algorithm submodule, fault detection submodule and data recording submodule.

[0020] The communication submodule provides a unified standard input and output interface protocol to realize the input and output interface between the collaborative control module and the outside world, including inputting mission information, inputting the status information of the node, inputting the sensor detection information of the node, inputting the status information of the neighboring node, inputting the target perception information of the neighboring node, outputting the status information of the node, outputting the target perception information of the node, and outputting the collaborative control instruction information.

[0021] The input mission information is the mission instructions planned in the mission preparation phase, including the communication topology configuration of multiple UUVs, the target operation area, and the target mission, and is output to the mission analysis submodule. It adopts a unified standardized instruction set, has universality and standardization characteristics, and can be used for the unified issuance and efficient analysis of multiple heterogeneous UUV mission instructions; The input node status information is the navigation status information of the module carrier UUV, including UTC timestamp, position, speed, attitude and other key information used for collaborative control calculation; The sensor detection information input to this node is the detection data received by the sonar detection sensor carried by the module carrier UUV, including key information such as target angle, signal-to-noise ratio and intensity for target state solution and tracking; The input neighbor node status information is the navigation status information of its neighboring UUV node, including UTC timestamp, position, speed, attitude and other key information used for collaborative control calculation; The input neighbor node target perception information is the target information obtained by the adjacent UUV node using its own perception information processing submodule to filter and process the sensor detection data and estimate the state, including target speed, target heading, target relative angle, target relative distance and other key information used for UUV behavior planning and control.

[0022] The output node status information is the input information of the neighboring node, including the timestamp, position, speed, posture and other key information of the node used for collaborative control calculation; The output target perception information of this node is the input information of the neighboring node, including the target detection angle, target solution position, target solution speed, target solution heading and confidence of this node after optimization by the perception information processing submodule, and other key information used for collaborative target detection and tracking.

[0023] The output collaborative control command information is the control command information of the UUV calculated by the collaborative control, including the expected speed, navigation and depth information that the UUV body needs to track. It adopts a unified standardized command set, has universality and standardization characteristics, and can realize the parsing and tracking of control commands of heterogeneous UUV platforms; The mission analysis submodule analyzes and outputs the mission information output by the communication submodule based on a unified standard mission instruction set. The submodule supports two modes of mission analysis, including a static mission mode and a dynamic mission mode.

[0024] The perception information processing submodule receives the task instruction information from the collaborative algorithm submodule to determine the execution task mode and the target perception information of the neighboring nodes and the sensor detection information of this node outputted from the communication submodule. First, according to the sensor detection information of this node, the sliding time window and the least square method are used to fit and filter the detection angle to obtain the filtered detection angle and the confidence of the angle estimation. Finally, the particle filter algorithm based on the boundary constraint is used to settle the motion elements of the detected target to obtain the estimated target speed, heading, relative distance and state estimation confidence. Secondly, combined with the target perception information of the neighboring nodes, the perception information with high confidence is selected as the credible data output. If the perception information of the neighboring nodes is selected, part of the data needs to be further converted, such as the relative angle of the neighboring node's target needs to be converted into the angle of the target relative to the body in the carrier coordinate system of this node. Finally, the above-processed perception information is output to the behavior decision submodule for further behavior planning control.

[0025] The behavior decision submodule inputs the perception information and task instruction information output by the perception information processing submodule to determine the task mode. For static tasks, the UUV is controlled by related behavior planning according to the preset process, while for dynamic tasks, the UUV is controlled by behavior planning based on three behavior strategies, including target search behavior decision, target tracking decision based on pure bearing and target tracking decision based on motion behavior analysis. The strategy switching is determined and selected according to the current state of the UUV, and finally the behavior control instruction information such as carrier speed and heading angle is output to the collaborative control algorithm submodule.

[0026] The target search behavior decision, that is, when the detection angle information and the calculated target motion elements in the perception information output by the perception information processing submodule are invalid for a certain period of time, it means that the target is lost and the target cannot be located. It is necessary to adopt an S-shaped maneuvering search method to increase the search detection range and re-search and locate the target. Figure 5 shown.

[0027] The target tracking decision based on pure bearing, that is, when the target motion element solved in the perception information output by the perception information processing submodule is invalid, the approximate direction of the target is determined only based on the detection angle information, and a virtual point 50 meters away is set as the tracking position point with this direction as the axis, and the UUV motion control is planned to move quickly in the suspected direction of the target, so as to get close to the target to obtain more effective information and complete a better tracking task.

[0028] The target tracking decision based on motion behavior analysis, that is, when all the perception information in the perception information processing submodule is valid, the target position information is determined according to the solved target relative angle and relative distance, so as to quickly plan the UUV motion control to achieve dynamic target tracking.

[0029] The specific planning method of the above-mentioned target tracking decision method based on pure bearing or motion behavior analysis is as follows: Figure 6 As shown. After obtaining the virtual tracking point or the actual tracking point, the global reference path is first generated and optimized. In order to meet the turning radius constraint, the Dubins curve is used to connect the start and end points, that is, the current position of the UUV is the starting point, and the calculated UUV virtual tracking point or actual tracking point is the end point. At the same time, the global reference path is optimized using cubic spline interpolation. After the global reference path is determined, it is necessary to project its current position onto the global reference path, and combine its own state information to generate multiple local sampling trajectories in the horizontal and vertical directions. The sampling points are connected by quintic polynomials to ensure trajectory continuity and dynamic smoothness. The local trajectory finally generated selects the optimal path through a multi-dimensional comprehensive evaluation and outputs control command information such as heading angle and speed. Refer to the following 5 evaluation indicators: 1) Smoothness cost :Low smoothness will cause additional lateral acceleration, which will cause the UUV to shake. Therefore, in order to reduce the shaking, a lateral offset secondary guide along the reference line is selected. The integral of the square is used as a criterion for the smoothness of the trajectory as follows.

[0030] ;

[0031] 2) Consistency Cost : Sudden changes in trajectory require excessive energy and additional control. Therefore, in order to avoid generating trajectories that are significantly different from previous trajectories, a measure of the similarity between previous trajectories and the current candidate path is needed. Here it is simplified to evaluate the current trajectory number i and the previous desired trajectory The absolute value of the difference between them is as follows.

[0032] ;

[0033] 3) Reference line cost : This criterion guides the UUV to follow a global reference path in order to position itself on the desired path. Since candidate paths are generated around this center by a defined lateral offset, the cost of following the reference trajectory is proportional to this offset. The farther the lateral offset is, the more the cost increases, given by: ; in is the lateral distance between trajectory i and the reference line.

[0034] 4) Security costs : To ensure the safe driving of UUV, the occupation grid within a certain time range of the lateral distance is mainly considered for safety assessment. If no collision occurs, the cost is 0, and the obstacles within a lateral distance around it are calculated.

[0035] ; When there are no obstacles around the UUV within a specific lateral distance range (obstacle-free), the safety cost is 0, indicating that the UUV is in a safe state and does not need to pay an additional "cost" to avoid collision.

[0036] When the horizontal distance When there are obstacles in the past, it is necessary to determine the relative position of the obstacle and the UUV (by The calculation method is ,in is the lateral distance parameter between the obstacle and the UUV at the i-th sampling point at a certain moment, and is a constant. As it approaches 0 (i.e., the closer the UUV is to the obstacle), the safety cost will gradually increase and approach 2. near When the safety cost is gradually reduced, it approaches 0. This can encourage the UUV to stay away from obstacles as much as possible when planning the path, ensuring its own safe driving.

[0037] In the actual UUV motion planning algorithm, the safety cost will be involved in the comprehensive evaluation of the path together with other costs (such as smoothness cost, consistency cost, reference line cost, tracking cost, etc.). By calculating and comparing the various costs of different paths, the path with the lowest total cost is selected as the driving path of the UUV, so as to ensure the safe operation of the UUV while ensuring the tracking target. For example, in the process of selecting the optimal path based on multi-dimensional comprehensive evaluation, the safety cost and other costs will be weighted and summed according to different weights to obtain the total cost function of each path (such as), and finally determine the optimal path and output the corresponding heading angle and speed and other control command information to the UUV carrier for tracking control.

[0038] 5) Track costs :In order to keep tracking the target, the calculation of tracking cost will be considered. The tracking cost mainly includes the distance error with the target and the navigation error: ; ; In the formula, is the longitudinal length of the entire reference line, is the longitudinal length of the projection point of the current UUV on the reference path, To maintain the vertical distance between expectations and goals. is the relative angle between UUV and target, is the current heading of the UUV.

[0039] Since the above different cost standards are measured in different units, it is necessary to first make the above indicators dimensionless based on normalization, as shown in the following formula: ; represents the value of the i-th cost index after normalization, is the original i-th cost index, The minimum value of this cost indicator among all the paths to be evaluated. is the maximum value, Finally, the total cost function of each trajectory is composed of the weighted sum of the normalized cost indicators, which is used to evaluate and select the optimal tracking trajectory, and output control command information such as heading angle and speed to the UUV carrier for tracking control. The total cost function formula is as follows: ; is the smoothness cost , safety costs , Tracking distance cost , Tracking Angle Cost The weight coefficient of is the normalized reference line cost The weight coefficient of is the consistency cost The weight coefficient of represents the total cost function of the i-th trajectory; The collaborative algorithm submodule is an integration of collaborative control algorithms. It mainly combines the relevant historical key data information in the data recording submodule, processes the task instruction information output by the current mission analysis submodule, the node status information received from the UUV carrier in the communication submodule, the status information of each neighbor node of the underwater acoustic communication, the behavior control instructions output by the behavior decision submodule, etc., outputs unified standard control instructions to the communication submodule, and transmits them to the UUV carrier bottom control unit to realize UUV motion control.

[0040] The fault detection and processing submodule receives the error information of each submodule and performs comprehensive processing, including the connection failure error of the communication submodule, the task instruction parsing error of the mission parsing submodule, the key fault detection information of the collaborative algorithm submodule, etc. This module determines whether the UUV has a fault and its specific type by analyzing the above information, and generates a unified standard fault processing control instruction to send to the collaborative algorithm submodule, and stores the key fault data information in the data recording submodule to provide data support for subsequent fault analysis and processing.

[0041] The data recording submodule is responsible for storing the key data of the system in the log file to support subsequent data analysis and processing, so as to facilitate operators to trace the task execution process and conduct problem analysis; at the same time, this submodule is also the public data area of ​​the system, and the key data stored in it is used for other submodules to call and store, which effectively reduces the resource occupation of each submodule for independently storing data and improves the system operation efficiency.

[0042] The command and control personnel formulate the mission task through the host computer software on the surface mother ship / shore-based command and control center, and the host computer software generates a mission task text file with a standard format based on the unified standard task instructions and sends it to each UUV node. The mission file contains UUV group information and topological relationships, supporting centralized topology and distributed topology. In addition, it also contains the mission mode, relevant mission task information such as key path points, and mission completion recovery area information.

Claims

1. A distributed heterogeneous multi-underwater unmanned vehicle collaborative control module for target detection and tracking, wherein the collaborative control module is installed in a sealed cabin of a UUV platform and is powered by a UUV power management module. The collaborative control module exchanges data with a UUV platform communication module and a shore-based platform through a unified standard communication protocol and does not participate in the UUV bottom-level control. Multiple UUVs are each equipped with a collaborative control module to form a distributed multi-UUV collaborative control system, characterized in that: The collaborative control module includes: The communication submodule provides a unified standard input and output interface for exchanging information with the UUV carrier and shore-based platform; Mission parsing submodule, which parses mission instructions based on a unified standard instruction set and supports static and dynamic mission modes; The perception information processing submodule receives and processes target detection data from the sonar sensor, including target angle, signal-to-noise ratio, and intensity; The behavior decision submodule determines the behavior strategy of the UUV based on the processed perception information and task instructions; The collaborative algorithm submodule executes the control algorithm and generates unified standard UUV control instructions; the fault detection and processing submodule monitors system errors and generates fault processing instructions; The data recording submodule stores key system data for other submodules to call and store to support subsequent analysis.

2. The distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking according to claim 1, characterized in that: The communication submodule comprises: 2.1 Input mission information: including the communication topology configuration of multiple UUVs, target operation area and target mission, and output to the mission analysis submodule; 2.2 Input the node status information, the module carrier UUV's own navigation status information, including UTC timestamp, position, speed, attitude and other key information, for collaborative control calculation; 2.3 Input the sensor detection information of this node: the detection data received by the sonar detection sensor carried by the module carrier UUV, including key information such as target angle, signal-to-noise ratio and intensity, which is used for target state solution and tracking; 2.4 Input neighbor node status information: The navigation status information of the neighboring UUV nodes, including UTC timestamp, position, speed, attitude and other key information, is used for collaborative control calculation; 2.5 Input the target perception information of neighboring nodes: The neighboring UUV nodes use their own perception information processing submodules to filter and process the sensor detection data and estimate the target information, including key information such as target speed, target heading, target relative angle and target relative distance, which are used for UUV behavior planning and control; 2.6 Output the status information of this node: the input information of neighboring nodes, including the timestamp, position, speed, attitude and other key information of this node, which is used for collaborative control calculation; 2.7 Output the target perception information of this node: the input information of neighboring nodes, including the key information such as target detection angle, target solution position, target solution speed, target solution heading and confidence after optimization by the perception information processing submodule of this node, which is used for collaborative target detection and tracking; 2.8 Output collaborative control command information: The control command information of the UUV calculated by collaborative control includes the expected speed, navigation and depth information that the UUV body needs to track. A unified standardized command set is used to realize the parsing and tracking of control commands of heterogeneous UUV platforms.

3. The distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking according to claim 1, characterized in that: The mission analysis submodule includes: 3.1 Static mission mode: used for area coverage search of static targets. The execution process has been fully pre-planned before the mission begins, that is, each stage of the mission and the execution order are fixed; 3.2 Dynamic mission mode: used for dynamic target search and tracking tasks, etc., which is dynamically determined based on real-time environmental data or changes in mission target status, and adjusted in real time based on actual conditions.

4. The distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking according to claim 1, characterized in that: The perception information processing submodule includes: receiving task instruction information from the collaborative algorithm submodule to determine the execution task mode, receiving target perception information of neighboring nodes and detection information of the node sensor from the communication submodule, fitting and filtering the detection angle using the sliding time window and least squares method, obtaining the detection angle and the confidence of the detection angle, estimating the target speed, heading, relative distance and the confidence of the state estimation based on the boundary constraint particle filter algorithm, selecting the perception information with high confidence as the trusted data after fusing the perception information of neighboring nodes, and passing the processed perception information to the behavior decision submodule for behavior planning and control.

5. The distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking according to claim 1, characterized in that: The behavioral decision submodule includes: The behavior decision submodule receives the perception information received from the perception information processing submodule, selects an appropriate behavior strategy according to the current state of the UUV, and determines whether the executed task is static or dynamic. When the executed task is a static task, behavior planning is performed according to a preset process, and at least one of the following behavior strategies is adopted: target search behavior decision, which is used to relocate the target when the target is lost; target tracking decision based on pure bearing, which is used to track when the resolution of target motion elements fails; target tracking decision based on motion behavior analysis, which is used to track when the resolution of target motion elements is valid.

6. The distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking according to claim 1, characterized in that: The collaborative algorithm submodule, combined with the relevant historical key data information in the data recording submodule, processes the task instruction information output by the current mission analysis submodule, the node status information received from the UUV carrier in the communication submodule, the status information of each neighboring node of the underwater acoustic communication, the behavior control instructions output by the behavior decision submodule, etc., outputs unified standard control instructions to the communication submodule, and transmits them to the bottom control unit of the UUV carrier to realize UUV motion control.

7. The distributed heterogeneous multi-underwater unmanned vehicle cooperative control module for target detection and tracking according to claim 4, characterized in that: The perception information processing submodule further converts the neighbor node perception information into data in the coordinate system of the current node when necessary.

Images