Distributed high-fidelity simulation system for large-scale high-speed AUVs based on real disturbance data

Abstract

A distributed high-fidelity simulation system for large-scale high-speed AUVs based on real disturbance data is proposed. Through a large-scale AUV virtual test vehicle subsystem, the system simulates the motion state of a real large-scale high-speed AUV physical prototype under the interaction of rigid body inertia, gravity, buoyancy, inertial hydrodynamic forces, centripetal force, Coriolis force, drag, thrust, rudder force, and external environmental disturbances. Collision attributes are added to the prototype through a physics engine. The system also utilizes an underwater 3D virtual environment subsystem to render a large-scale underwater environment, build dynamic and static obstacle visualization models, and simulate the generation of realistic ocean currents and wave disturbances with high fidelity. The simulation presents a realistic underwater environment for a large, high-speed AUV. A motion solution subsystem describes the transformation relationships between the fixed coordinate systems of all objects in the underwater virtual environment and solves the motion equations of the large AUV, synchronously updating its pose and velocity states. A data management subsystem manages various data generated during the simulation, supporting write and read data operations, and also manages and coordinates data communication between nodes, supporting receive and send data operations. A visualization subsystem loads and dynamically displays the underwater 3D environment and the large AUV virtual test vehicle model, synchronously updating real-time data curves and visualizing the AUV's motion trajectory.

Description

Technical Field

[0001] This invention relates to a technology in the field of underwater robots, specifically a high-fidelity three-dimensional visualization simulation system for performance testing and algorithm verification of large-scale high-speed underwater robots based on real ocean current and wave disturbance data. Background Technology

[0002] The performance testing and intelligent algorithm (environmental perception, planning and decision-making, control algorithm, etc.) of large high-speed autonomous underwater vehicles (AUVs) in actual sea areas require a lot of manpower, financial resources and material resources. Moreover, the operation and testing environment has great uncertainty, such as ocean currents and waves, making it difficult to ensure the safety of personnel and operating equipment. Summary of the Invention

[0003] This invention addresses the challenges of high difficulty in R&D and algorithm verification for large-scale high-speed AUV systems, poor code portability across platforms, and low fidelity in simulation systems. It proposes a distributed high-fidelity simulation system for large-scale high-speed AUVs based on real disturbance data. This system can test the open-loop performance of AUVs in a virtual ocean environment with realistic ocean currents and wave disturbances, and also verify the reliability and effectiveness of artificial intelligence-related algorithms, reducing the difficulty of system R&D and algorithm verification. Furthermore, the distributed architecture facilitates loose coupling of multiple hardware systems, enabling cross-platform code portability and testing. It allows for dynamic visualization of the underwater virtual environment, the virtual test vehicle for large-scale high-speed AUVs, and virtual sensors, as well as real-time data and motion trajectory visualization. The high-fidelity simulation of real underwater test scenarios improves human-machine communication efficiency, makes results more intuitive, and allows users to observe and analyze the motion state of large-scale AUVs from multiple perspectives, further enhancing the efficiency and flexibility of hardware system development and algorithm research.

[0004] This invention is achieved through the following technical solution:

[0005] This invention relates to a distributed high-fidelity simulation system for large-scale high-speed AUVs based on real disturbance data, comprising: a large-scale AUV virtual test vehicle subsystem, an underwater 3D virtual environment subsystem, a motion solution subsystem, a data management subsystem, and a visualization subsystem. The large-scale AUV virtual test vehicle subsystem simulates the motion state of a real large-scale high-speed AUV physical prototype on a scale, under the interaction of rigid body inertia, gravity, buoyancy, inertial hydrodynamic forces, centripetal force, Coriolis force, drag, thrust, rudder force, and external environmental disturbances, and adds collision attributes to the prototype through a physics engine. The underwater 3D virtual environment subsystem renders and simulates the motion of a large-scale underwater environment, including static and dynamic simulations. The system constructs a visualization model of the underwater obstacle and generates realistic ocean currents and wave disturbances to simulate the real underwater environment of a large, high-speed AUV with high fidelity. The motion solution subsystem describes the transformation relationship between the fixed coordinate systems of all objects in the underwater virtual environment and solves the motion equations of the large AUV, synchronously updating its pose and velocity states. The data management subsystem manages various data generated during the simulation, supports writing and reading data operations, and can also manage and coordinate data communication issues between nodes, supporting receiving and sending data operations. The visualization subsystem loads and dynamically displays the underwater 3D environment and the virtual test vehicle model of the large AUV, synchronously updating real-time data curves and visualizing the AUV's motion trajectory.

[0006] The large-scale AUV virtual test vessel subsystem includes: a motion association module, a motion perception module, a motion decision module, and a motion control module. Specifically: the motion association module uses linkages and joints as basic units to represent the topological relationships of the various components of the large-scale AUV; the motion perception module converts data returned by the virtual sensor components into dynamic environmental information and state information during the high-speed navigation of the large-scale AUV, and outputs this information to the motion decision module; the motion decision module plans strategies (such as path planning or control decisions) based on the information fed back by the motion perception module and cooperates with the motion control module to complete the designated tasks; the motion control module receives instructions from the motion decision module, responds quickly, calculates actuator actions, and accurately and with high quality completes the desired actions.

[0007] The aforementioned topological relationships include: the appearance texture and geometric collision model of each component of the large AUV, as well as the physical properties of mass and inertia; the joints associate the upper and lower links according to the parent-child relationship, and the parent link can be the child of other links at the same time, and similarly, the child link can be the parent of other links at the same time.

[0008] The virtual sensor component refers to a sensor device that is abstracted through a simulation system and has the same or similar function as an actual physical sensor, and its relevant attribute parameters can be set.

[0009] The sensors include: an odometer for measuring the voyage distance of a large AUV; a virtual camera for acquiring underwater image information; a single-beam sonar for acquiring the distance from the nearest obstacle to the sensor; a multi-beam sonar for acquiring and returning a set of distance data; a depth gauge for acquiring the operating depth data of the large AUV; an IMU for acquiring the position and attitude data of the AUV relative to the world coordinate system; a GPS for acquiring the absolute position data of the AUV; a DVL for acquiring the height of the AUV above the seabed and its speed data; and a force sensor for acquiring force and torque data.

[0010] The specified tasks include: underwater navigation, autonomous obstacle avoidance, target tracking, multi-body cooperation, or a combination thereof.

[0011] The motion control module includes a position controller, an attitude controller, and a speed controller. It responds quickly to the instructions issued by the motion decision module of the large AUV and completes the desired action accurately and with high quality.

[0012] The position controller takes a reference position as input and data collected by the position sensor on the AUV as feedback, and drives the large AUV to the reference position by adjusting the thruster speed and rudder angle. The attitude controller takes a reference attitude as input and data collected by the attitude sensor on the AUV as feedback, and drives the large AUV to the reference attitude by adjusting the rudder angle. The speed controller takes a reference speed as input and the current speed of the AUV as feedback, and drives the large AUV to the reference speed by adjusting the thruster speed.

[0013] The appearance texture and geometric collision model are constructed using SolidWorks and 3D MAX to create a visual model and a geometric model with a collision mesh. The visual model has shape, color, material, and texture attributes.

[0014] The large AUV virtual test subsystem is based on a conventional single rigid body AUV, with an enlarged length-to-diameter ratio. Its main body is a single rigid body structure, and the stern is equipped with a tunnel thruster and a cross rudder. The cross rudder includes a vertical rudder and a horizontal rudder. The pitch and yaw attitudes of the AUV can be adjusted by swinging the rudder.

[0015] The underwater three-dimensional virtual environment subsystem includes: a seabed model with collision physical properties, dynamic and static object models, and real ocean disturbances. The seabed model includes flat land, seamounts, and trench topography; static objects include shipwrecks and rusty iron pillars; dynamic objects include swimming fish, underwater vehicles, and suspended objects; and real ocean disturbances include the disturbance forces and moments of ocean currents and waves.

[0016] The seabed and dynamic / static object models were constructed using SolidWorks, 3D MAX, and World Machine software, and corresponding geometric collision models were established based on the simplified models, thereby reducing the computational load of the physics engine in solving complex environment collision models.

[0017] The motion solution subsystem includes a coordinate system transformation solution module and a large AUV motion equation solution module. The coordinate system transformation solution module performs coordinate system transformation based on the position and velocity information in one coordinate system to obtain the position and velocity information in another coordinate system. The large AUV motion equation solution module performs dynamic calculation based on the AUV actuator command information to obtain the AUV's pose and velocity state information.

[0018] The coordinate transformation solution module includes: dynamic and static coordinate system transformation solution units, which are designed to calibrate objects in the underwater environment using coordinate systems. Specifically, the dynamic coordinate system transformation solution unit calibrates moving objects, that is, it describes the coordinate transformation relationship between the fixed coordinate system of the moving object and the reference system (optional), such as the transformation relationship between the body coordinate system of a large AUV and the world coordinate system, the transformation relationship between the fixed coordinate system of the rudder and the body coordinate system, etc.; the static coordinate system transformation solution unit describes the coordinate transformation relationship between different fixed reference systems, such as the transformation between the fixed coordinate system of the seabed model and the world coordinate system, etc.

[0019] The large AUV motion equation solving module includes a kinematic unit and a dynamic unit. It uses ODE45 to solve the motion equations and synchronously updates the state information of the large AUV. The kinematic unit describes the state representation of the large AUV in the world coordinate system, and the dynamic unit describes the changes in the velocity state of the large AUV relative to its own fixed coordinate system under the action of hydrodynamics, environmental disturbance forces and actuator forces.

[0020] The aforementioned state representation refers to the following: the aircraft is an ideal rigid body, and all mechanical actions acting on it are equivalent to a single external force (moment); the inertial frame defined on Earth is unaffected by the Earth's rotation. Under these assumptions, based on Euler's first and second laws, the rigid body dynamics equations of the aircraft in the body coordinate system are established as follows: Where: v is velocity; M is the derivative of v, representing acceleration; RB C is the rigid body inertial mass matrix; RB (v) is the Coriolis force-centripetal force matrix; τ RB It is the sum and vector of external forces (torques), including inertial hydrodynamic forces, resistance, gravity, buoyancy, lift, rudder force, propulsion, external environmental disturbance forces and their respective torques.

[0021] The data management subsystem includes a data read / write module and a communication management module. The data read / write module performs data writing and reading operations. Specifically, the write operation stores sensor data, large AUV pose and velocity status data, and timestamp data in .bag format files to an external memory for later offline data analysis and processing. The read operation retrieves relevant data from the 3D visualization simulation system in the external memory and supports data playback, i.e., reproducing the dynamic image simulation process corresponding to the data in the visualization subsystem. The communication management module manages and allocates the nodes (processes) of the Robot Operating System (ROS). The implementation of a subsystem consists of one or more nodes.

[0022] The aforementioned 3D visualization simulation system is based on the communication architecture of the ROS system. The system consists of multiple nodes, each of which is executed as an independent executable file (usually an executable file compiled from C++ or a Python script) and is responsible for a single function, such as dynamic solution, motion decision and control. The numerous nodes are managed and coordinated in a distributed manner through the communication management module.

[0023] The nodes communicate with each other using a publish-subscribe topic approach. Specifically, during node communication, both publisher and subscriber nodes need to register in the communication management module. Publishers then publish topics, and subscribers, under the guidance of the communication management module, subscribe to relevant topics and process received messages, thus establishing communication between subscribers and publishers. Message routing does not require direct connections between nodes; the same topic can have multiple subscribers and multiple publishers.

[0024] The visualization subsystem includes: an underwater environment configuration module, a main camera module, a window configuration module, and a data visualization module. The underwater environment configuration module sets parameters such as underwater environment color, visibility, color attenuation, and maximum water depth. The main camera module captures images within its field of view, such as underwater scenes, sea level, and sky, and displays them in the window. The window manager module configures functions such as window aspect ratio, resolution, shader on / off, trajectory display, and Bullet physics engine on / off. The data visualization module displays the AUV's real-time trajectory, curves showing the change of relevant state quantities over time, and real-time sensor data.

[0025] Technical effect

[0026] Compared to existing technologies, this invention employs a distributed architecture, facilitating loose coupling between multiple hardware systems. This means the code can be ported to other hardware systems without modification. By considering the collision properties of objects, virtual sensors enable information interaction between the large-scale AUV virtual test vessel and the virtual environment, providing reliable environmental information data for the research and effectiveness verification of intelligent algorithms such as environmental perception, motion decision-making, and control. Compared to existing technologies, this invention lays the foundation for the future engineering applications of intelligent algorithms, providing a reference for the structural design and optimization of real AUVs and the development of software and hardware systems. Furthermore, it allows for planning and controller parameter tuning before field testing, improving the safety of personnel and equipment during field testing. Attached Figure Description

[0027] Figure 1 This is a visualization of the high-fidelity simulation system of the present invention;

[0028] Figure 2 This is a schematic diagram of the system of the present invention;

[0029] Figure 3 This is a diagram showing the relationships between the models in the 3D visualization simulation system of this invention.

[0030] Figure 4 This is a schematic diagram of the motion coordinate system of the large AUV virtual test vessel model of the present invention. Detailed Implementation

[0031] like Figure 1 As shown in this embodiment, a distributed high-fidelity simulation system for large-scale high-speed AUVs based on real disturbance data is included. The system comprises: a large-scale AUV virtual test vehicle subsystem, an underwater 3D virtual environment subsystem, a motion solution subsystem, a data management subsystem, and a visualization subsystem. Specifically, the large-scale AUV virtual test vehicle subsystem includes a motion association module, a motion perception module, a motion decision module, and a motion control module; the underwater 3D virtual environment subsystem includes seabed topography, dynamic and static objects, and real ocean currents and wave disturbances; the motion solution subsystem includes a coordinate system transformation solution module and a large-scale AUV motion equation solution module; the data management subsystem includes a data read / write module and a communication management module; and the visualization subsystem includes an underwater environment configuration module, a main camera module, a window configuration module, and a data visualization module.

[0032] The models in the large AUV virtual test vehicle subsystem and the underwater 3D virtual environment subsystem are loaded into the visualization subsystem in OBJ format via XML format files to achieve model visualization. Sensor data from the motion perception module, control input data from the motion control module, and state data of various objects in the underwater 3D virtual environment in the large AUV virtual test vehicle subsystem flow into the data management subsystem for data storage and transmission. The data management subsystem and the motion solving subsystem perform bidirectional data transmission. The motion solving subsystem reads the AUV control input and the state data of various underwater objects from the data manager, calculates the AUV motion state, and the fixed coordinate system transformation relationship of various underwater objects, and then transmits this data to the data management subsystem. The data management subsystem then transmits the relevant data to the visualization subsystem, driving model movement and achieving data visualization.

[0033] In this embodiment, the large AUV virtual test vehicle subsystem is based on a conventional single rigid-body AUV. The model is scaled arbitrarily with the same length-to-diameter ratio by setting the AUV length, and the dynamic parameters of the corresponding AUV size are automatically configured simultaneously, with the speed set to 14 knots or higher. The structure of the large AUV virtual test vehicle in the preferred embodiment of the 3D simulation system is as follows: Figure 4 As shown.

[0034] The motion association module describes the topological relationship between the hull and each actuator. The hull is a single rigid body structure. The actuators include a vertical rudder, a horizontal rudder, and a propeller. The vertical rudder and the horizontal rudder are installed orthogonally at the rear of the hull, and the propeller is installed at the stern of the hull.

[0035] The motion sensing module describes a large AUV equipped with various sensing devices for sensing the environment and its own state.

[0036] The models in the large AUV virtual test vehicle subsystem and the underwater three-dimensional virtual environment subsystem are built using SolidWorks, 3D Studio Max and World Machine software, and exported as OJB format files with attributes such as texture, material, and geometric collision. They are described in URDF and XML formats and displayed in the visualization subsystem.

[0037] like Figure 3 As shown, the coordinate system transformation and solving module of the three-dimensional visualization simulation system describes the transformation relationship between the fixed coordinate systems of various objects in the underwater environment. The objects in this embodiment include various seabed topography, dynamic and static objects, AUV hull and its components. The coordinate systems of all objects are connected directly or indirectly, and there should be no disconnection. Only by complying with the above constraints can the coordinate system transformation and solving module work normally.

[0038] like Figure 4As shown, the motion equation solving module solves the motion equations of the large AUV and updates its motion state in real time. The motion equations include kinematic equations and dynamic equations. The body coordinate system in the figure is fixed at the center of the hull and together with the world coordinate system, it meets the East-North-Geocentric system standard. The mathematical symbols in the motion equations are described in Table 1.

[0039] Table 1. Mathematical Symbol Description of Equations of Motion

[0040]

[0041] The kinematic equations are established as follows: the position vector of the origin of the body coordinate system relative to the world coordinate system is η = [X,Y,Z,φ,θ,ψ]. T Where X, Y, Z represent position coordinates, φ, θ, ψ represent attitude angular coordinates, [ ] T The square brackets [] represent the transpose of the coordinate system; the velocity vector of the origin of the body coordinate system in the body coordinate system is ν = [u,v,w,p,q,r]. T Where u, v, w represent linear velocities, and p, q, r represent angular velocities. The transformation relationship between the velocity at the origin of the body coordinate system and the world coordinate system satisfies: Where: Jacobian transformation matrix J(η)∈R 6×6 for s(·), c(·), and t(·) represent sin(·), cos(·), and tan(·), respectively.

[0042] The aforementioned dynamic equations are established as follows. Before establishing the dynamic equations of a large AUV, the following assumptions must be defined: the large AUV is an ideal rigid body, and all mechanical actions acting on it are equivalent to a single external force (moment); the world coordinate system defined on Earth is not affected by the Earth's rotation. Under the above assumptions, according to Euler's first and second laws, the rigid body dynamic equations of the large AUV are: Where: the first three equations are translational equations, the last three equations are rotational equations, and m is the mass of the large AUV. Represents the inertia tensor matrix, [x G y G z G ] T The coordinates of the center of gravity of a large AUV.

[0043] The above expression can be rearranged into a vector expression as follows: in: τ is the derivative of v, representing acceleration; RB M is the sum vector of external forces (moments); RB Let be the rigid body inertial mass matrix, which is constant over time and satisfies: C R B(v) is the Coriolis force-centripetal force matrix, which is a skew-symmetric matrix and satisfies: Large AUVs are subjected to external forces (torques) τ RB =τ H +τ E +τ C , where: τ H =τ A +τ D +g+τ l This represents the force (moment) acting on a fluid, including inertial hydrodynamic forces (moment) τ. A Fluid resistance (moment) τ D Lift (torque) τ l And restoring force (torque) g; τ E τ represents environmental disturbance forces (moments), such as waves and ocean currents; C It represents the control force (torque), including the thrust of the propeller and the force and torque generated by the rudder.

[0044] According to Kirchhoff's equations, the inertial hydrodynamic force caused by the added mass effect can be calculated as follows: Where: M A Let C represent the additional mass matrix. A (v) represents the Coriolis force-centripetal force matrix caused by the added mass effect; the fluid resistance experienced by the large AUV is: τ D =C D (v)v, where: C D (v) represents the linear and nonlinear fluid damping matrices.

[0045] When the center of gravity and center of buoyancy of an AUV do not coincide and are not equal in magnitude, the resultant force of gravity and buoyancy in the body coordinate system is: Where: G and B represent the gravity and buoyancy of the AUV, respectively, and r g r b These are the vector representations of the center of gravity and center of buoyancy in the body coordinate system, respectively; the lift force experienced by a large AUV includes the lift force τ acting on the main body. bl and the lift τ on the rudder fl Two parts: τ l =τ bl +τ fl .

[0046] In summary, the dynamic equations governing large AUVs in underwater environments are as follows: This can be further simplified to: Where: M = M RB -M A C(v) = C A (v)-C RB (v).

[0047] Finally, by using ODE45 to solve the above kinematic and dynamic equations, the real-time state variables of the large AUV can be obtained.

[0048] The communication management module enables communication between nodes through topics. It receives data from the large AUV virtual test vehicle subsystem and the underwater 3D virtual environment subsystem, and conducts bidirectional data communication with the motion solution subsystem. Then, all data is published to the visualization subsystem as topics for display. The data read / write module uses the rosbag command to package all topic messages into a .bag format file and stores it on an external hard drive for later offline data analysis and processing, improving data security and reusability. Similarly, this module can also read relevant data from the 3D simulation system on the external hard drive using the rosbag command and supports data playback, allowing the simulation process corresponding to the data to be reproduced in the visualization subsystem.

[0049] The nodes include: motion equation solving nodes, motion decision nodes, motion control nodes, motion sensing nodes, and visualization nodes. During node communication, nodes must register in the communication management module before they can publish and subscribe to relevant topics. These topics include: AUV control variables, AUV state variables, various sensor-related topics, reference path topics, and coordinate transformation-related topics. Each node, under the management of the communication management module, can simultaneously publish and subscribe to multiple topics.

[0050] The visualization subsystem is implemented based on the UWSim simulator and Qt toolkit, and includes an underwater environment configuration module, a main camera module, a window management module, and a data visualization module. The first three are plugins for the UWSim simulator, configured using XML documents validated with DTD documentation. Specifically, the underwater environment configuration module configures important parameters such as underwater color, color attenuation, and maximum water depth; the main camera settings module sets parameters such as whether the main camera follows the target object, the field of view angle, aspect ratio, and the closest / farthest shooting distance; and the window management module manages window display attributes, including resolution, enabling / disabling the renderer, world coordinate system, enabling AUV real-time trajectory display, and enabling the Bullet physics engine.

[0051] like Figure 1As shown, the data visualization module is used to display the real-time trajectory, motion state variables, and real-time sensor data of the AUV in a window. Specifically, the real-time trajectory visualization of the AUV can be achieved through the configuration window management module; the visualization of the motion state variables and real-time sensor data can be achieved by subscribing to relevant topics through the rqt_plot node in the Qt toolbox, and then displaying the curves of the relevant data changing over time in the plotting window in real time.

[0052] During the simulation experiment, the key parameters of the large AUV were designed as follows: AUV length 14 meters, diameter 2 meters, speed 14 knots, initial depth 100 meters; marine environment configuration as follows: seabed topography 10000m × 10000m, underwater color configuration dark blue, color attenuation set to no attenuation, maximum water depth set to 200 meters; main camera module configuration: tracking AUV, field of view 80 degrees, aspect ratio 1.33, minimum shooting distance 1 meter, maximum shooting distance 200 meters; window management module configuration as follows: resolution set to 1600 pixels × 1200 pixels, image renderer enabled, world coordinate system set to East-North-Ground coordinate system, AUV real-time trajectory display enabled, Bullet physics engine enabled; in sensor parameter configuration, multibeam sonar parameters were configured according to the actual sensor with a maximum detection range of 200 meters and a detection angle ±60 degrees. The actual simulation effect is as follows. Figure 1 As shown.

[0053] The above-described specific implementations can be partially adjusted by those skilled in the art in different ways without departing from the principles and purpose of the present invention. The scope of protection of the present invention is defined by the claims and is not limited to the above-described specific implementations. All implementation schemes within the scope of the claims are bound by the present invention.

Claims

1. A distributed high-fidelity simulation system for large-scale high-speed AUVs based on real disturbance data, characterized in that, include: The system comprises a large-scale AUV virtual test vehicle subsystem, an underwater 3D virtual environment subsystem, a motion solution subsystem, a data management subsystem, and a visualization subsystem. Specifically, the large-scale AUV virtual test vehicle subsystem simulates the motion of a real large-scale high-speed AUV physical prototype on a scale, under the interaction of rigid body inertia, gravity, buoyancy, inertial hydrodynamics, centripetal force, Coriolis force, drag, thrust, rudder force, and external environmental disturbances, and adds collision attributes to the prototype through a physics engine. The underwater 3D virtual environment subsystem renders a large-scale underwater environment, builds dynamic and static obstacle visualization models, and utilizes realistic ocean currents... The simulation generates high-fidelity wave disturbances to simulate the real underwater environment of a large, high-speed AUV; the motion solution subsystem describes the transformation relationships between the fixed coordinate systems of all objects in the underwater virtual environment and solves the motion equations of the large AUV, synchronously updating its pose and velocity states; the data management subsystem manages various data generated during the simulation, supports writing and reading data operations, and can also manage and coordinate data communication between nodes, supporting receiving and sending data operations; the visualization subsystem loads and dynamically displays the underwater 3D environment and the large AUV virtual test vehicle model, synchronously updating real-time data curves and visualizing the AUV's motion trajectory. The large-scale AUV virtual test subsystem is based on a conventional single-rigid-body AUV, with an enlarged length-to-diameter ratio. Its main body is a single-rigid-body structure, equipped with a tunnel thruster and a cross-shaped rudder at the stern. The cross-shaped rudder includes a vertical rudder and a horizontal rudder; by swinging the rudder, the pitch and yaw attitudes of the AUV can be adjusted. This large-scale AUV virtual test subsystem includes: a motion association module, a motion perception module, a motion decision module, and a motion control module. Specifically: the motion association module uses links and joints as basic units to represent the topological relationships of the various components of the large-scale AUV; the motion perception module converts data returned by the virtual sensor components into dynamic environmental information and state information during the high-speed navigation of the large-scale AUV, and outputs it to the motion decision module; the motion decision module plans strategies based on the information fed back by the motion perception module and cooperates with the motion control module to complete the designated task; the motion control module receives instructions from the motion decision module, responds quickly, calculates actuator actions, and accurately and with high quality completes the desired action; The aforementioned topological relationships include: the appearance texture and geometric collision model of each component of the large AUV, as well as the physical properties of mass and inertia; the joints associate the upper and lower links according to the parent-child relationship, and the parent link can simultaneously serve as the child of other links, and similarly, the child link can simultaneously serve as the parent of other links.

2. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 1, characterized in that, The virtual sensor component refers to a sensor device that is abstracted by a simulation system and has the same or similar function as an actual physical sensor, and its relevant attribute parameters can be set. The specified tasks include: underwater navigation, autonomous obstacle avoidance, target tracking, multi-body cooperation, or a combination thereof.

3. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 2, characterized in that, The motion control module includes a position controller, an attitude controller, and a speed controller, wherein: based on the instructions issued by the motion decision module received by the large AUV, the desired action is quickly completed; The position controller takes a reference position as input and data collected by the position sensor on the AUV as feedback, and drives the large AUV to the reference position by adjusting the thruster speed and rudder angle. The attitude controller takes a reference attitude as input and data collected by the attitude sensor on the AUV as feedback, and drives the large AUV to the reference attitude by adjusting the rudder angle. The speed controller takes a reference speed as input and the current speed of the AUV as feedback, and drives the large AUV to the reference speed by adjusting the thruster speed.

4. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 1, characterized in that, The underwater three-dimensional virtual environment subsystem includes: a seabed model with collision physical properties, dynamic and static object models, and real ocean disturbances. The seabed model includes flat land, seamounts, and trench topography; static objects include shipwrecks and rusty iron pillars; dynamic objects include swimming fish, underwater vehicles, and suspended objects; and real ocean disturbances include the disturbance forces and moments of ocean currents and waves.

5. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 1, characterized in that, The motion solution subsystem includes a coordinate system transformation solution module and a large AUV motion equation solution module. The coordinate system transformation solution module performs coordinate system transformation based on the position and velocity information in one coordinate system to obtain the position and velocity information in another coordinate system. The large AUV motion equation solution module performs dynamic calculation based on the AUV actuator command information to obtain the AUV's pose and velocity state information.

6. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 5, characterized in that, The coordinate system transformation solution module includes: dynamic and static coordinate system transformation solution units, which are designed to calibrate objects in the underwater environment using coordinate systems. Specifically, the dynamic coordinate system transformation solution unit calibrates moving objects, that is, it describes the coordinate transformation relationship between the fixed coordinate system and the reference system of the moving object, and the transformation relationship between the fixed coordinate system of the rudder and the body coordinate system; the static coordinate system transformation solution unit describes the coordinate transformation relationship between different fixed reference systems. The large AUV motion equation solving module includes a kinematic unit and a dynamic unit. It uses ODE45 to solve the motion equation and synchronously updates the state information of the large AUV. The kinematic unit describes the state representation of the large AUV in the world coordinate system; the dynamic unit describes the change of the velocity state of the large AUV relative to its own fixed coordinate system under the action of hydrodynamics, environmental disturbance forces and actuator forces. The aforementioned state representation refers to the following: the aircraft is an ideal rigid body, and all mechanical forces acting on it are equivalent to a single external torque; the inertial frame defined on Earth is unaffected by Earth's rotation; and according to Euler's first and second laws, the rigid body dynamics equations of the aircraft in the body coordinate system are established as follows: ,in: For speed; for The derivative of represents acceleration; The inertial mass matrix of the rigid body; The Coriolis force-centripetal force matrix; It is the sum and vector of external torques, including inertial hydrodynamic forces, resistance, gravity, buoyancy, lift, rudder force, propulsion, external environmental disturbance forces and their respective torques.

7. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 1, characterized in that, The data management subsystem includes a data read / write module and a communication management module. Specifically, the data read / write module performs data writing and reading operations. Specifically, the writing operation stores sensor data, large AUV pose and velocity status data, and timestamp data in .bag format files to an external memory for later offline data analysis and processing. The reading operation reads relevant data from the 3D visualization simulation system in the external memory and supports data playback, i.e., reproducing the dynamic image simulation process corresponding to the data in the visualization subsystem. The communication management module manages and allocates the nodes of the robot operating system. The implementation of a subsystem consists of one or more nodes.

8. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 7, characterized in that, The aforementioned 3D visualization simulation system is based on the communication architecture of the ROS system. The system consists of multiple nodes, each of which is executed as an independent executable file and is responsible for a single function. The numerous nodes are managed and coordinated in a distributed manner through the communication management module.

9. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 7, characterized in that, The nodes communicate with each other by publishing and subscribing to topics. Specifically, during node communication, publisher nodes and subscriber nodes need to register in the communication management module. Then, the publisher publishes topics, and the subscribers subscribe to relevant topics under the command of the communication management module. After receiving messages, the subscribers process the messages, thereby establishing communication between the subscriber and the publisher. Message routing does not require direct connections between nodes, and the same topic can have multiple subscribers or multiple publishers.

10. The distributed large-scale high-speed AUV high-fidelity simulation system based on real disturbance data according to claim 1, characterized in that, The visualization subsystem includes: an underwater environment configuration module, a main camera module, a window configuration module, and a data visualization module. The underwater environment configuration module sets parameters such as underwater environment color, visibility, color attenuation, and maximum water depth. The main camera module captures images within its field of view and displays them in the window. The window manager module configures the window aspect ratio, resolution, shader on / off settings, trajectory display, and Bullet physics engine on / off functionality. The data visualization module displays the AUV's real-time trajectory, curves showing the change of relevant state variables over time, and real-time sensor data.

Images