A method and system for docking an unmanned surface vehicle with a docking station

Abstract

The application provides a docking method and system of an unmanned surface vehicle and a docking station, and relates to the field of autonomous docking of unmanned surface vehicles. The method comprises: when the distance between the unmanned surface vehicle and the docking station is less than a set value, acquiring state data of both sides respectively; performing data fusion on real-time state data and predicted state data calculated based on historical data to generate target optimal state data; constructing a gravitational field with the front end of the docking station as a gravitational source and a repulsive field with the rear end of the docking station and obstacles as repulsive sources according to the target optimal state data; calculating the relative position of the unmanned surface vehicle and the docking station; determining a control mode according to the relative position, generating a virtual force field in combination with the gravitational field, the repulsive field and the mode algorithm; and generating a motion path according to the direction and size of the virtual force field and controlling docking. Thus, through the cooperative acquisition of double-end state data, the docking deviation is effectively reduced; in combination with the path planning strategy of dynamic mode switching, the obstacle avoidance success rate is significantly improved, thereby improving the docking efficiency.

Description

Technical Field

[0001] This application relates to the field of autonomous docking of unmanned surface vessels, and in particular to a docking method and system for unmanned surface vessels and docking stations. Background Technology

[0002] Unmanned surface vehicles (USVs) are increasingly used in marine monitoring and security patrols, but their autonomous docking with docking stations faces numerous challenges in complex marine environments. Current technologies typically employ single-end positioning for docking, but this method cannot effectively eliminate docking station sway or displacement errors, leading to significant docking deviations. Furthermore, traditional static path planning algorithms suffer from low obstacle avoidance success rates, large path oscillation amplitudes, and a high rate of local minima in complex scenarios, easily affecting docking efficiency and stability. Summary of the Invention

[0003] This application provides a docking method and system for an unmanned surface vessel and a docking station, so as to at least solve the above-mentioned technical problems existing in the prior art.

[0004] According to a first aspect of this application, a method for docking an unmanned surface vessel with a docking station is provided, the method comprising:

[0005] When the distance between the unmanned surface vessel and the docking station is less than a first set distance, the status data of the unmanned surface vessel and the docking station are acquired respectively. The status data includes coordinates, attitude, acceleration and angular velocity.

[0006] The status data of the unmanned surface vessel and the docking station, as well as the predicted status data of the unmanned surface vessel and the docking station, are fused to generate target status data of the unmanned surface vessel and the docking station. The predicted status data is obtained based on the historical status data of the unmanned surface vessel and the docking station.

[0007] Based on the target status data of the unmanned surface vessel and the docking station, a gravitational field with the front end of the docking station as the gravitational source and a repulsive field with the rear end of the docking station and obstacles as the repulsive source are constructed.

[0008] Calculate the relative positions of the unmanned surface vessel and the docking station based on the target status data of the unmanned surface vessel and the docking station;

[0009] Based on the relative positions of the unmanned surface vessel and the docking station, the control mode of the unmanned surface vessel is determined, and a virtual force field of the unmanned surface vessel is generated based on the gravitational field, repulsive field, control mode and its corresponding algorithm.

[0010] Based on the direction and magnitude of the virtual force field, the motion path of the unmanned surface vessel is determined, and the unmanned surface vessel is controlled to dock with the docking station according to the motion path of the unmanned surface vessel.

[0011] In one possible implementation, predicted state data is obtained based on historical state data of the unmanned surface vessel and the docking station, including:

[0012] Based on the historical state data from the previous set time, the angular velocity and acceleration are shown, and the intermediate predicted state data for the current time is predicted by integrating the angular velocity of the inertial measurement unit.

[0013] The intermediate predicted state data and historical state data are fused to obtain the predicted state data for the current moment.

[0014] In one possible implementation, data fusion is performed on the status data of the unmanned surface vessel and the docking station, as well as the predicted status data of the unmanned surface vessel and the docking station, including:

[0015] Obtain the state error covariance corresponding to the predicted state data of the unmanned surface vessel and the docking station;

[0016] The global coordinates in the status data of the unmanned surface vessel and docking station are converted into Cartesian coordinates in the local Cartesian coordinate system. The global coordinates include longitude, latitude and elevation.

[0017] The heading angle shown by the converted Cartesian coordinates and state data is used as the actual observation vector, and the observation equation is constructed based on the actual observation vector to obtain the deviation. The deviation is used to show the degree of deviation between the predicted state data and the actual observation vector.

[0018] Calculate the Kalman gain based on the state error covariance, the observation matrix, and the preset measurement noise covariance;

[0019] Based on the Kalman gain, the bias, and the predicted state data of the unmanned surface vessel and the docking station, the optimal state data of the target is calculated, and the optimal state data of the target is used as the predicted state data for the next set time.

[0020] In one possible implementation, the repulsive field includes a first repulsive field and a second repulsive field; correspondingly,

[0021] Based on the target state data of the unmanned surface vessel and the docking station, a gravitational field with the front end of the docking station as the gravitational source and a repulsive field with the rear end of the docking station and obstacles as repulsive sources are constructed, including:

[0022] Using the coordinates of the docking station's front end, as shown in the target status data of the unmanned surface vessel and docking station, as the center of the gravity source, a gravity field is constructed. This gravity field is used to determine gravity and its direction based on the gravity function.

[0023] Using the coordinates of the rear end of the docking station shown in the target status data of the unmanned surface vessel and the docking station as the center of the repulsive force source, a first repulsive force field is constructed. The first repulsive force field is used to determine the first repulsive force and the direction of the first repulsive force according to the first repulsive force function.

[0024] The coordinates of all obstacles on the unmanned surface vessel are obtained, and a second repulsive field is constructed based on the coordinates of all obstacles. The second repulsive field is used to determine the second repulsive force and the direction of the repulsive force based on the second repulsive force function.

[0025] In one possible implementation, the relative position includes the relative distance, relative azimuth, and relative heading deviation between the unmanned surface vessel and the docking station; based on the relative position of the unmanned surface vessel and the docking station, the control mode of the unmanned surface vessel is determined, including:

[0026] If the relative distance between the unmanned surface vessel and the docking station meets the preset contact range and the relative heading deviation between the unmanned surface vessel and the docking station is less than the set deviation range, the control mode is determined to be the forward alignment mode.

[0027] If the relative azimuth angle between the unmanned surface vessel and the docking station is greater than the set azimuth angle and the relative distance between the unmanned surface vessel and the docking station is less than a set multiple of the length of the docking station, the control mode is determined to be the rearward bypass mode.

[0028] If the relative heading deviation between the unmanned surface vessel and the docking station is greater than the set deviation range, the control mode of the unmanned surface vessel is determined to be the alignment mode.

[0029] In one possible implementation, a virtual force field for the unmanned surface vessel is generated based on the gravitational field, repulsive field, control mode, and corresponding algorithm, including:

[0030] In the forward alignment control mode, a gravitational vector is calculated based on the gravitational function of the gravitational field, a first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and a second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the weight of the gravitational vector using a gravitational line alignment algorithm, it is superimposed with the first and second repulsive vectors to obtain a resultant force vector. A virtual force field is constructed based on the resultant force vector.

[0031] In the control mode of rearward bypass mode, a gravitational vector is calculated based on the gravitational function of the gravitational field, a first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and a second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the weight of the first repulsive vector using a repulsive point guidance algorithm, it is superimposed with the gravitational vector and the second repulsive vector to obtain a resultant force vector. A virtual force field is constructed based on the resultant force vector.

[0032] In the alignment control mode, a gravitational vector is calculated based on the gravitational function of the gravitational field, a first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and a second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the lateral weight and forward weight of the gravitational vector using the gravitational line alignment algorithm, it is superimposed with the first repulsive vector and the second repulsive vector to obtain a resultant force vector. A virtual force field is constructed based on the resultant force vector.

[0033] In one possible implementation, the motion path of the unmanned surface vessel is determined based on the direction and magnitude of the virtual force field, including:

[0034] The path point generation rules are determined based on the direction and magnitude of the virtual force field, and the generation rules include the spacing and direction of the path points.

[0035] Within each control cycle, the next path point is generated based on the corresponding generation rules and the direction and magnitude of the virtual force field;

[0036] By connecting a series of path points in sequence, the motion path of the unmanned surface vessel is generated.

[0037] In one possible implementation, controlling the unmanned surface vessel to dock with the docking station according to the unmanned vessel's motion path includes:

[0038] Based on the motion path, calculate the required acceleration and angular velocity of the unmanned surface vessel at each point on the path, and generate corresponding motion commands.

[0039] The motion command is decomposed into thruster differential speed and rudder angle control quantities, and anti-saturation optimization is performed in combination with the unmanned surface vessel dynamics model;

[0040] The optimized motion commands are sent to the unmanned surface vessel, driving the unmanned vessel's motors and servos to perform the docking action with the docking station.

[0041] According to a second aspect of this application, a docking system for an unmanned surface vessel and a docking station is provided, the system comprising:

[0042] The unmanned surface vessel is equipped with a dual-antenna system and an inertial measurement unit to acquire its own status data.

[0043] The docking station is equipped with a dual-antenna system and an inertial measurement unit to acquire its own status data; the control system is deployed on the unmanned surface vessel and includes a processor and a memory communicatively connected to the processor. The memory stores instructions that can be executed by the processor, and the instructions are executed by the processor to enable the processor to perform the above-mentioned methods.

[0044] An inter-device communication system is used to control data communication between the unmanned surface vessel and the docking station through a communication server and encryption algorithms;

[0045] A remote monitoring communication system for monitoring data from the communication server and the front end via a communication server and encryption algorithms.

[0046] In one possible implementation, the unmanned surface vessel is also equipped with a lidar for detecting obstacles during the docking process between the unmanned surface vessel and the docking station, and for triggering an emergency stop when the distance to the docking station is detected to be less than a second preset distance.

[0047] The unmanned surface vessel is also used for dynamic interception via a command verification module; the dynamic interception via the command verification module includes: when the speed is detected to be greater than a first set speed or the angular velocity is greater than a first set angular velocity, the speed is limited by the command over-limit interception function of the command verification module.

[0048] The docking method and system for an unmanned surface vessel (USV) and a docking station disclosed in this application acquires state data of both the USV and the docking station when the distance between them is less than a first predetermined distance. This state data includes coordinates, attitude, acceleration, and angular velocity. The system then fuses the state data of the USV and the docking station with predicted state data to generate target optimal state data for both the USV and the docking station. The predicted state data is obtained based on historical state data of the USV and the docking station. Finally, the system determines the target optimal state of the USV and the docking station. Based on the state data, a gravitational field is constructed with the front end of the docking station as the gravitational source, and a repulsive field is constructed with the rear end of the docking station and obstacles as repulsive sources. The relative positions of the unmanned surface vessel (USV) and the docking station are calculated based on their optimal target state data. The control mode of the USV is determined based on their relative positions, and a virtual force field for the USV is generated based on the gravitational field, repulsive field, control mode, and their corresponding algorithms. The motion path of the USV is determined based on the direction and magnitude of the virtual force field, and the USV is controlled to dock with the docking station according to this motion path. Thus, by collaboratively acquiring state data from both ends, docking deviation is effectively reduced; combined with a path planning strategy that uses dynamic mode switching, obstacle avoidance success rate is significantly improved, thereby increasing docking efficiency.

[0049] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0050] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of this application are illustrated in the drawings by way of example and not limitation, in which:

[0051] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts.

[0052] Figure 1 This paper illustrates a schematic diagram of the composition of a docking system between an unmanned surface vessel and a docking station, as provided in an embodiment of this application.

[0053] Figure 2 A schematic diagram illustrating the implementation process of the docking method between an unmanned surface vessel and a docking station provided in an embodiment of this application is shown.

[0054] Figure 3 This paper illustrates a schematic diagram of the force field construction operation of the docking method between an unmanned surface vessel and a docking station provided in an embodiment of this application. Detailed Implementation

[0055] To make the objectives, features, and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0056] Figure 1 This paper presents a schematic diagram of the composition of the docking system between the unmanned surface vessel and the docking station provided in an embodiment of this application.

[0057] refer to Figure 1This application provides a docking system for an unmanned surface vessel (USV) and a docking station. The system includes: an USV equipped with a dual-antenna system and an inertial measurement unit (IMU) for acquiring its own status data; a docking station equipped with a dual-antenna system and an IMU for acquiring its own status data; a control system deployed on the USV, including a processor and a memory communicatively connected to the processor, the memory storing instructions executable by the processor, which are then executed to enable the processor to perform a docking method between the USV and the docking station; an inter-device communication system (not shown), used to complete data communication between the USV and the docking station via a communication server and encryption algorithms; and a remote monitoring communication system (not shown), used to complete data monitoring between the communication server and the front-end via a communication server and encryption algorithms. The docking station is a berthing point for the USV and may include, but is not limited to, a docking platform, a communication relay station, an energy supply station, or an emergency shelter.

[0058] Specifically, to achieve coordinate synchronization and matching between the two ends, both the unmanned surface vessel (USV) and the docking station are equipped with an SGEllipse-D dual-antenna GPS RTK (Real-Time Kinematic) system. This system acquires three-dimensional coordinates in the WGS-84 (World Geodetic System-1984) coordinate system in real time, updating at a frequency of 10Hz. The specific information includes longitude, latitude, and elevation. The USV performs Euclidean distance matching based on the pre-set docking station coordinates and calculates the relative distance by measuring the coordinate difference. When the relative distance is less than 50 meters, the docking method of the unmanned surface vessel and the docking station of this application is triggered.

[0059] During the docking process between the unmanned surface vessel (USV) and the docking station, the USV's GPS and RTK dual-antenna system provides centimeter-level real-time positioning, including information in three dimensions: longitude, latitude, and altitude. Simultaneously, it uses an IMU (Inertial Measurement Unit) to measure attitude, angular velocity, and acceleration to compensate for dynamic errors. The docking station also continuously transmits its precise status information, including coordinates, attitude, acceleration, and angular velocity, via GPS, RTK, and IMU for the USV's use. Both sides synchronize data through secure wireless communication, ensuring sub-meter-level relative positioning in complex environments and achieving stable closed-loop control, ultimately completing a high-precision autonomous docking.

[0060] In terms of communication, the entire system is divided into two parts: inter-device communication and remote monitoring communication. During inter-device communication, the docking station uses AES-256 (Advanced Encryption Standard-256) encrypted UDP (User Datagram Protocol) to send an identity packet containing the device's MAC address (Media Access Control Address, 6 bytes, or 48 hexadecimal bits), a timestamp (accurate to milliseconds), and a digital signature. After receiving this information, the unmanned surface vessel calculates a hash value (256 bits, or 32 bytes, hexadecimal string) using the SHA-256 (Secure Hash Algorithm-256) hash algorithm to verify the docking station's identity and prevent malicious impersonation from interfering.

[0061] Simultaneously, the USV and the docking station communicate via WebSocket, transmitting their respective data to the ROS2 (Robot Operating System 2) deployed on the unmanned surface vessel, i.e., the control system. The transmitted data includes the real-time attitude (position and orientation), motion state (angular velocity and acceleration) of the USV and the docking station, as well as environmental data (such as obstacle information). At the same time, the USV receives path planning and control commands from ROS2 via WebSocket.

[0062] During remote monitoring and communication, WebSocket (wss: / / ) encrypted with TLS (Transport Layer Security) version 1.2 / 1.3 is used, with the underlying AES-256-GCM (Advanced Encryption Standard-256 bit Galois / Counter Mode) algorithm to ensure secure transmission of WebSocket communication. Simultaneously, a browser can connect to the WebSocket server and subscribe to the ROS2Topic. The server pushes ROS2 data to the front end in real-time in JSON (JavaScript Object Notation), displaying information such as the USV trajectory (latitude and longitude sequence) and docking station status (dynamically updated coordinates and heading angle). This facilitates better monitoring of whether the process is running normally. Communication interruption protection is also enabled during communication; if no new instructions are received for more than 100ms, the USV enters hover mode.

[0063] In one embodiment of this application, the unmanned surface vessel is further equipped with a lidar for detecting obstacles during docking with the docking station and triggering an emergency stop when the distance to the docking station is detected to be less than a second preset distance. The unmanned surface vessel is also used for dynamic interception via a command verification module. The dynamic interception via the command verification module includes limiting the speed when the detected speed is greater than a first preset speed or the angular velocity is greater than a first preset angular velocity, by using the command over-limit interception function of the command verification module.

[0064] Specifically, the unmanned surface vessel is equipped with a LiDAR (Light Detection and Ranging) system to trigger an emergency stop when the distance to the docking station is detected to be less than a second preset distance (e.g., 10 cm). Simultaneously, the unmanned surface vessel has a command over-limit interception function: when the detected speed exceeds a first preset speed (e.g., 1.3 m / s) or angular velocity exceeds a first preset angular velocity (e.g., 0.8 rad / s), it automatically limits the speed. The second preset distance, the first preset speed, or the angular velocity can be configured according to actual conditions, and this application does not impose specific limitations.

[0065] It should be noted that the description of the system in this application embodiment is similar to the description of the method embodiment below, and has similar beneficial effects as the method embodiment, therefore it will not be repeated. For any technical details not covered in the docking system of the unmanned surface vessel and docking station provided in this application embodiment, they can be understood based on the following description of the docking method of the unmanned surface vessel and docking station.

[0066] Figure 2 The diagram illustrates the implementation flow of the docking method between an unmanned surface vessel and a docking station provided in an embodiment of this application.

[0067] refer to Figure 2 This application provides a method for docking an unmanned surface vessel with a docking station, the method comprising:

[0068] Operation 101: When the distance between the unmanned surface vessel and the docking station is less than a first set distance, acquire the status data of the unmanned surface vessel and the docking station respectively. The status data includes coordinates, attitude, acceleration and angular velocity.

[0069] The USV performs Euclidean distance matching in real time based on the pre-set docking station coordinates (latitude and longitude of the docking station). It calculates the relative distance through the coordinate difference. When the relative distance is less than the first set distance (e.g., 50m), the initial docking process is triggered, and the status data of both are acquired separately.

[0070] The status data includes coordinates, attitude, acceleration, and angular velocity. Coordinates are three-dimensional coordinates in the WGS-84 Global Positioning System (GPS) acquired by the unmanned surface vessel (USV) and docking station via the SBG Ellipse-D dual-antenna GPS RTK system, including longitude, latitude, and elevation information. Attitude refers to the roll, pitch, and yaw angles measured by the IMU (Integrated Device Unit) of the USV and docking station, reflecting their tilt and turning states. Acceleration is the linear acceleration measured by the IMU, including acceleration information in the x, y, and z directions. Angular velocity is the attitude angular velocity measured by the IMU, including angular velocity information about the x, y, and z axes.

[0071] Operation 102 involves fusing the state data of the unmanned surface vessel and the docking station with the predicted state data of the unmanned surface vessel and the docking station to generate the target optimal state data of the unmanned surface vessel and the docking station. The predicted state data is obtained based on the historical state data of the unmanned surface vessel and the docking station.

[0072] To generate more accurate status data, the real-time status data of the unmanned surface vessel and docking station are fused with the predicted status data to generate the optimal status data of the target.

[0073] In one embodiment of this application, the method of obtaining predicted state data based on historical state data of unmanned surface vessels and docking stations includes: predicting intermediate predicted state data for the current moment by integrating the angular velocity and acceleration shown in the historical state data of the previous set time through the inertial measurement unit; and fusing the intermediate predicted state data with the historical state data to obtain the predicted state data for the current moment.

[0074] Specifically, based on historical state data of the unmanned surface vessel and docking station, a certain algorithm is used to predict the current state data. The historical state data refers to the state data at the previous set point in time.

[0075] In one embodiment of this application, the data fusion method can be either the Extended Kalman Filter (EKF) algorithm or the Particle Filter (PF) algorithm.

[0076] Operation 103: Based on the target optimal state data of the unmanned surface vessel and the docking station, construct a gravitational field with the front end of the docking station as the gravitational source and a repulsive field with the rear end of the docking station and obstacles as the repulsive source.

[0077] To guide the unmanned surface vessel (USV) safely and efficiently to the docking station and avoid collisions, a virtual force field environment needs to be provided for the USV, specifically including:

[0078] Constructing a gravitational field: Design a gravitational field with the front end of the docking station as the gravitational source. The gravitational direction always points towards the front end of the docking station, and the magnitude varies with distance. This is used to attract unmanned surface vessels to move towards the docking station.

[0079] Constructing a repulsive field: Using the rear end of the docking station as the repulsive source, a repulsive field is constructed with its direction away from the rear end and its magnitude varying with distance. A Gaussian distribution repulsive field is generated for each obstacle. The repulsive field is used to repel unmanned surface vessels and avoid collisions.

[0080] In this way, the USV is guided to move towards the docking station by the gravitational field, while the repulsive field is used to avoid collisions with obstacles, thus improving the safety and effectiveness of path planning.

[0081] Operation 104: Calculate the relative positions of the unmanned surface vessel and the docking station based on the target optimal state data of the unmanned surface vessel and the docking station.

[0082] Accurate relative position information is fundamental for determining the control mode and generating the virtual force field; therefore, it is also necessary to calculate the relative position of the unmanned surface vessel (USV) and the docking station. Specifically, based on the target optimal state data of the USV and the docking station, the relative position between them is calculated. This relative position includes relative coordinates (Δx, Δy, Δz), relative distance (calculated using the Euclidean distance formula), relative azimuth angle, and relative heading deviation.

[0083] Operation 105: Based on the relative positions of the unmanned surface vessel and the docking station, determine the control mode of the unmanned surface vessel, and generate a virtual force field for the unmanned surface vessel based on the gravitational field, repulsive field, control mode and its corresponding algorithm.

[0084] To address the issues of path oscillation, local minima, and environmental adaptability in complex dynamic environments encountered by existing static path planning algorithms, this application employs a dynamic path planning method. Based on the different relative positions of the USV and the docking station, different control modes are activated, such as forward alignment and rearward detour. This allows for real-time adjustment of the gravitational and repulsive fields to generate a virtual force field that guides the USV's movement, ensuring a smooth docking process. This effectively reduces path oscillation amplitude, decreases the incidence of local minima, adapts to complex marine environments, and improves the success rate and efficiency of docking.

[0085] Operation 106: Determine the motion path of the unmanned surface vessel based on the direction and magnitude of the virtual force field, and control the unmanned surface vessel to dock with the docking station based on the motion path of the unmanned surface vessel.

[0086] To achieve precise USV docking, the guidance of the virtual force field needs to be translated into specific motion paths and control commands. Specifically, based on the direction and magnitude of the virtual force field, starting from the current USV position, the next path point is generated along the direction of the virtual force field. By continuously iterating and updating the path points, a dynamically smooth motion path is formed. Finally, the unmanned surface vessel is controlled to move along the motion path to achieve docking with the docking station.

[0087] Thus, this embodiment of the application accurately acquires and integrates the state data of the unmanned surface vessel (USV) and the docking station, constructs a dynamic virtual force field to guide path planning, and adopts different control modes to ensure smooth docking, thereby achieving high-precision and high-reliability autonomous docking, effectively improving path smoothness and safety, and reducing the incidence of path oscillation and local minima.

[0088] In one embodiment of this application, the above-described operation 102 involves data fusion of the state data of the unmanned surface vessel and the docking station, as well as the predicted state data of the unmanned surface vessel and the docking station. This includes: obtaining the state error covariance corresponding to the predicted state data of the unmanned surface vessel and the docking station; converting the global coordinates in the global coordinate system of the state data of the unmanned surface vessel and the docking station into Cartesian coordinates in the Cartesian local coordinate system, where the global coordinates include longitude, latitude, and elevation; using the converted Cartesian coordinates and the heading angle shown in the state data as the actual observation vector, and constructing an observation equation based on the actual observation vector to obtain the deviation, wherein the deviation is used to indicate the degree of deviation between the predicted state data and the actual observation vector; calculating the Kalman gain based on the state error covariance, the observation matrix, and a preset measurement noise covariance; and calculating the target optimal state data based on the Kalman gain, the deviation, and the predicted state data of the unmanned surface vessel and the docking station, and using the target optimal state data as the predicted state data for the next set time.

[0089] Specifically, the data fusion process can be viewed as:

[0090] 1) Obtain the state error covariance of the predicted state data. This variance is used to describe the uncertainty and error distribution of the predicted state data and reflects the degree of deviation between the predicted state data and the actual situation.

[0091] 2) Convert the coordinate format in the state data to a set unified format. For example, if the original coordinates are global coordinates in a global coordinate system, they can be converted to Cartesian coordinates in a Cartesian (East-North-Up) local coordinate system to facilitate subsequent calculations and processing, so that coordinate data from different sources can be fused in the same reference system.

[0092] 3) The heading angle shown by the Cartesian coordinates after the conversion format and the state data is used as the actual observation vector, and the observation equation is constructed based on the actual observation vector to obtain the deviation. The observation equation is used to show the mathematical relationship between the actual measured state data and the predicted state data, and the deviation is used to show the degree of deviation between the predicted state data and the actual observation vector.

[0093] 4) Based on the state prediction bias, state error covariance, observation matrix and preset noise covariance, the Kalman gain is further calculated. The Kalman gain is used to show the correction strength of the actual measurement data to the predicted state during the fusion process.

[0094] 6) Based on the calculated Kalman gain, state prediction bias, and predicted state data, determine the optimal target state data after fusion.

[0095] Furthermore, to facilitate understanding of the data fusion process in this application, a specific application example is provided below. In this example, data fusion can be understood as performing EKF filtering on the data, specifically including:

[0096] 1) Initial parameter settings

[0097] Before starting the EKF filtering process, initial parameters are set, including the observation matrix, state transition matrix, process noise covariance, measurement noise covariance, and state error covariance at time 0. The state transition matrix is ​​determined by the motion model of the unmanned surface vessel, and the measurement noise covariance is pre-set through sensor calibration or empirical methods.

[0098] 2) Obtain the initial state

[0099] At time t=0, acquire the initial state information of the USV and the docking station, including position (x, y, z), attitude (roll, pitch, yaw), and linear velocity (v). x v y v z ) and angular velocity (ω) x ω y ω z ).

[0100] 3) Prediction phase

[0101] During the prediction phase, the pose x at the next moment is predicted by integrating the angular velocity and acceleration data from the IMU. k|k-1 The GPS coordinates (longitude, latitude, elevation) are converted to Cartesian coordinates (x, y, z) in the ENU local coordinate system, and the converted ENU coordinates are used as part of the actual observation vector. The observation vector includes the ENU coordinates and the GPSRTK heading angle ψ, i.e., y = z. k Construct the observation equation: yk =Hx k|k-1 +v k Where H is the observation matrix, used to map the state vector to the measurement space, v k The deviation reflects the degree of discrepancy between the predicted state and the actual measurement.

[0102] 4) Calculate the Kalman gain

[0103] Based on the predicted state error covariance P k|k-1 Given the observation matrix H and the measurement noise covariance R, calculate the Kalman gain Kk using the formula Kk. k =P k|k-1 H T HP k|k-1 H T +R) -1 Among them, P k|k-1 Through formula P k|k-1 =FP k-1|k-1 F T +Q is obtained during the prediction phase, where F is the state transition matrix and Q is the process noise covariance. Initially, P... 0|0 It is a unit array.

[0104] 5) Update Phase

[0105] During the update phase, the Kalman gain K is utilized. k Correcting the predicted state yields the optimal state estimate x after fusing the measurements. k|k The calculation formula is x k|k =x k|k-1 +K k v k Simultaneously, update the state error covariance P. k|k The calculation formula is P k|k =(IK k H)P k|k-1 .

[0106] Repeat steps 1)-5) above to continuously optimize the state estimation and output the optimal estimated state, i.e. the target optimal state data and the error covariance.

[0107] Thus, through the above EKF filtering process, the position accuracy can be improved to ±2cm and the attitude angle accuracy to ±0.3°, achieving centimeter-level docking requirements.

[0108] Figure 3 This paper illustrates a schematic diagram of the force field construction operation of the docking method between an unmanned surface vessel and a docking station provided in an embodiment of this application.

[0109] refer to Figure 3In one embodiment of this application, the repulsive field includes a first repulsive field and a second repulsive field. Operation 103, based on the target optimal state data of the unmanned surface vessel and the docking station, constructs a gravitational field with the front end of the docking station as the gravitational source and a repulsive field with the rear end of the docking station and obstacles as repulsive sources, including:

[0110] Operation 201: Using the coordinates of the docking station's front end shown in the target optimal state data of the unmanned surface vessel and docking station as the center of the gravity source, a gravity field is constructed. The gravity field is used to determine gravity and its direction based on the gravity function.

[0111] Specifically, with the front end of the docking station as the target point, a gravitational force is generated pointing towards the docking station. The gravitational field is designed with the center of the front end of the docking station as the gravitational source, and the gravitational function is as follows:

[0112]

[0113] in, k att Let q be the gravitational coefficient (typically configurable to 10 N / m), and q be the Cartesian coordinates of the unmanned surface vessel. q dock Using Cartesian coordinates for the docking station, the direction of gravity always points towards the target point.

[0114] Operation 202: Using the coordinates of the rear end of the docking station shown in the target optimal state data of the unmanned surface vessel and the docking station as the center of the repulsive force source, a first repulsive force field is constructed. The first repulsive force field is used to determine the first repulsive force and the direction of the first repulsive force according to the first repulsive force function.

[0115] By setting the rear end of the docking station as the repulsive force source, a first repulsive force field is constructed, and the corresponding first repulsive force function is:

[0116]

[0117] Among them, K rep The repulsion coefficient (configurable to 50 N / m), d (q) d0 is the distance between USV and the repulsion source, and d0 (configurable to 10m) is the radius of the repulsion force.

[0118] Operation 203: Obtain the coordinates of all obstacles on the unmanned surface vessel, and construct a second repulsive field based on the coordinates of all obstacles. The second repulsive field is used to determine the second repulsive force and the direction of the repulsive force based on the second repulsive force function.

[0119] Based on the LiDAR point cloud clustering results, i.e., the ENU coordinates and distance of each obstacle relative to the USV, a Gaussian distributed repulsive field (second repulsive field) is generated for each obstacle. The corresponding second repulsive function is:

[0120]

[0121] Where, k obs Let q be the repulsive force coefficient of the obstacle. obs σ represents the coordinates of the obstacle, and σ represents the range of influence of the obstacle (usually 5~10 meters).

[0122] In one embodiment of this application, in the above-mentioned operation 105, the relative position includes the relative distance, relative azimuth, and relative heading deviation between the unmanned surface vessel and the docking station; based on the relative position between the unmanned surface vessel and the docking station, the control mode of the unmanned surface vessel is determined, including:

[0123] If the relative distance between the unmanned surface vessel and the docking station meets the preset contact range and the relative heading deviation between the unmanned surface vessel and the docking station is less than the set deviation range, the control mode is determined to be the forward alignment mode.

[0124] Specifically, when the USV enters the preset contact range (usually 10-50°) in front of the docking station, and the relative heading deviation is less than the set deviation range (e.g., 15°), it indicates that the USV's heading is basically aligned with the docking station's axis, meeting the conditions for direct docking. Therefore, the control mode at this time is determined to be the forward alignment mode (FRONT), enabling the USV to smoothly approach and complete docking along the docking station's axis. The relative heading deviation refers to the angle between the USV's heading and the docking station's axis heading, which can be calculated from the heading angles shown by the USV and docking station attitudes.

[0125] If the relative azimuth angle between the unmanned surface vessel and the docking station is greater than the set azimuth angle and the relative distance between the unmanned surface vessel and the docking station is less than a set multiple of the length of the docking station, the control mode is determined to be the rearward bypass mode.

[0126] When the USV is located behind the docking station and the relative azimuth angle is greater than the set azimuth angle threshold (usually 180°±30°), and the relative distance is less than a set multiple of the docking station length (usually 3 times), the USV is in a relatively close area behind the docking station. If it continues to sail in a straight line, it may collide with the docking station or fall into a local minimum. Therefore, the control mode is set to REAR mode to guide the USV to detour to the front of the docking station to increase the success rate of docking.

[0127] When the relative heading deviation between the unmanned surface vessel and the docking station exceeds the set deviation range, the control mode of the unmanned surface vessel is determined to be the alignment mode.

[0128] When the relative heading deviation between the USV and the docking station exceeds the set deviation range (typically 15°), it indicates a significant deviation between the USV's heading and the docking station's axis. Directly docking along the normal path in this situation may lead to path oscillation or docking failure. Therefore, the control mode is set to ALIGN (alignment mode), allowing the USV to quickly adjust its heading and align with the docking station's axis, preparing for subsequent docking. Specifically, in alignment mode, if the relative heading deviation decreases to 10°, the system exits ALIGN mode and switches to FRONT mode.

[0129] In one embodiment of this application, in operation 105 above, generating a virtual force field for the unmanned surface vessel based on the gravitational field, repulsive field, control mode, and corresponding algorithm includes:

[0130] In the forward alignment control mode, the gravitational vector is calculated based on the gravitational function of the gravitational field, the first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and the second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the weight of the gravitational vector using the gravitational line alignment algorithm, it is superimposed with the first and second repulsive vectors to obtain the resultant force vector. A virtual force field is constructed based on the resultant force vector.

[0131] In the forward alignment mode, the algorithm used is the gravity line alignment algorithm, which forces the USV's course to align with the docking station's axis, reducing path oscillation amplitude. Specifically, the gravity line alignment algorithm can be:

[0132]

[0133] in, Let this be the USV heading normal vector. Let k be the USV heading unit vector. 11 k is the horizontal alignment factor (usually 0.8). 12 This is the forward thrust coefficient (usually 1.2).

[0134] Specifically, the virtual force field in this application is constructed based on the Artificial Potential Field Algorithm (APF). The principle is to calculate the "virtual force" (the vector sum of attraction and repulsion) acting on the USV in real time using artificial potential field functions (gravitational function and repulsion function), i.e., the virtual force field. The direction of the virtual force within the virtual force field corresponds to the target motion direction of the USV, and the magnitude of the resultant force corresponds to the target velocity of the USV. The construction process of the virtual force field is as follows:

[0135] 1) Potential Field Synthesis: The gravitational function outputs a gravitational vector, the first repulsive function outputs a first repulsive vector, and the second repulsive function outputs a second repulsive vector. Each force vector includes both direction and magnitude. The gravitational force increases with the distance between the unmanned surface vessel and the docking station. The gravitational vector can be determined by the gravitational coefficient k. att The intensity is adjusted; the closer the unmanned surface vessel is to the docking station, the greater the repulsive force. The repulsive force vector can be determined by the repulsive force coefficient k. rep The effective radius d0 limits the range of influence.

[0136] 2) Calculate the resultant force vector on the USV: gravitational force vector + first repulsive force vector + second repulsive force vector.

[0137] When the control mode is forward alignment mode, step 2) above, calculating the resultant force vector acting on the USV, can be considered as adjusting the weight of the gravity vector based on the relative distance and relative heading deviation using a gravity alignment algorithm, and increasing the lateral alignment coefficient k. 11 and forward thrust coefficient k 12 The adjusted gravitational vector is added to the first repulsive vector and the second repulsive vector to obtain the virtual force (resultant force vector), and the virtual force field corresponding to the virtual force is generated.

[0138] In the rearward bypass control mode, the gravitational vector is calculated based on the gravitational function of the gravitational field, the first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and the second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the weight of the first repulsive vector using the repulsive point guidance algorithm, it is superimposed with the gravitational vector and the second repulsive vector to obtain the resultant force vector. A virtual force field is constructed based on the resultant force vector.

[0139] When the control mode is rearward bypass mode, a repulsion guidance algorithm is used. The repulsion guidance algorithm is as follows:

[0140]

[0141] Where, k 21 The repulsion strength coefficient (usually 50 Nm) 2 ), k 22 This is the heading maintenance factor (usually 1.0).

[0142] Specifically, the virtual force field construction process in the current mode remains consistent with the aforementioned forward alignment mode, with only the resultant force vector calculation process being redesigned. In the rearward bypass mode, after calculating the gravitational vector, the first repulsive vector, and the second repulsive vector using the gravitational and repulsive functions, the process of calculating the resultant force vector can be regarded as adjusting the weight of the first repulsive vector through the repulsive guidance algorithm, increasing the repulsive intensity coefficient and the heading maintenance coefficient, and adding the adjusted gravitational vector with the first and second repulsive vectors to obtain the virtual force (resultant force vector), so that the USV maintains a safe distance from the rear of the docking station during the bypass process and gradually turns towards the front of the docking station.

[0143] In the parallel control mode, the gravitational vector is calculated based on the gravitational function of the gravitational field, the first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and the second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the lateral weight and forward weight of the gravitational vector using the gravitational line alignment algorithm, it is superimposed with the first repulsive vector and the second repulsive vector to obtain the resultant force vector. A virtual force field is constructed based on the resultant force vector.

[0144] Specifically, the virtual force field construction process in the current mode remains consistent with the aforementioned forward alignment mode, only the resultant force vector calculation process is redesigned, and the lateral alignment coefficient k of the gravity line alignment algorithm is changed. 11 and forward thrust coefficient k 12 The values ​​are swapped to increase the weight of the lateral gravity term by 50%, forcing the USV to align with the docking station axis. Forward thrust is reduced to decrease the USV's forward speed, preventing untimely course corrections due to high speed. In alignment mode, after calculating the gravity vector, the first repulsive vector, and the second repulsive vector using the gravity and repulsive functions, the process of calculating the resultant force vector can be viewed as adjusting the weight of the gravity vector using a gravity line alignment algorithm, increasing the lateral alignment coefficient k11 and the forward thrust coefficient k12, and adding the adjusted gravity vector to the first and second repulsive vectors to obtain the virtual force (resultant force vector), generating the virtual force field corresponding to the virtual force.

[0145] In one embodiment of this application, the repulsion coefficient k is adjusted according to the wave height during the docking process between the USV and the docking station. att and gravitational coefficient k rep .

[0146] Specifically, in practical applications, adjusting the attraction and repulsion coefficients according to different sea conditions (such as wave height) allows the USV to better adapt to environmental changes. For example, when the wave height is low (<0.5 meters), the default attraction and repulsion coefficients (1.0 times) can be used. When the wave height is high (0.5~1.0 meters or 1.0~1.5 meters), the repulsion coefficient can be appropriately increased. For wave heights of 0.5~1.0 meters, the repulsion coefficient can be adjusted to 1.2 times, and for wave heights of 1.0~1.5 meters, it can be adjusted to 1.5 times to enhance obstacle avoidance capabilities. Simultaneously, the attraction coefficient can be appropriately decreased. For wave heights of 0.5~1.0 meters, the repulsion coefficient can be adjusted to 0.9 times, and for wave heights of 1.0~1.5 meters, it can be adjusted to 0.8 times to prevent excessive attraction from causing instability in the waves.

[0147] In one embodiment of this application, the above operation 106, which determines the motion path of the unmanned surface vessel based on the direction and magnitude of the virtual force field, includes: determining the generation rules of path points based on the direction and magnitude of the virtual force field, wherein the generation rules include the spacing and direction of the path points; generating the next path point based on the corresponding generation rules and the direction and magnitude of the virtual force field within each control cycle; and connecting a series of path points in sequence to generate the motion path of the unmanned surface vessel.

[0148] Specifically, starting from the current USV position, the next path point is generated along the direction of the virtual force field. The distance of the next path point is determined by the target linear velocity shown by the virtual force field and the processing cycle (≤8ms). For example, when the linear velocity is 1m / s, a path point with a step size of 0.008m is generated every 8ms. Through continuous iteration (e.g., updating the virtual force field and path points every 8ms), these consecutive path points are chained together to form a dynamically smooth motion path, ensuring that the path oscillation amplitude is controlled within ±0.3m. Here, the direction of the virtual force field is the direction of the generation rule, and the distance of the next path point is the spacing of the generation rule.

[0149] In one embodiment of this application, controlling the docking of an unmanned surface vessel (USV) with a docking station based on the USV's motion path includes: calculating the required acceleration and angular velocity of the USV at each point on the path based on the motion path, and generating corresponding motion commands; decomposing the motion commands into thruster differential speed and rudder angle control quantities, and performing anti-saturation optimization in conjunction with the USV's dynamic model; and sending the optimized motion commands to the USV to drive the USV's motors and servos to perform the docking action with the docking station.

[0150] Specifically, for each path point in the motion path, the motion requirements corresponding to the path point are converted into motion commands including linear velocity and angular velocity (such as forward velocity along the resultant force direction and angular velocity for adjusting heading) using the APF algorithm. The motion commands are published in the form of ROS2Topic ( / cmd_vel topic) and transmitted to the USV in real time via TLS-encrypted WebSocket (wss: / / ). The USV is controlled based on an adaptive proportional-integral-derivative controller (PID) to decompose linear velocity and angular velocity into thruster differential (controlling steering) and rudder angle control (controlling forward direction). Finally, these are sent to the USV's Ardupilot autopilot via the MAVLink protocol, driving the motors and servos to execute the commands, complete path tracking, and thus achieve docking with the docking station.

[0151] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0152] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0153] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for docking an unmanned surface vessel with a docking station, characterized in that, The method includes: When the distance between the unmanned surface vessel and the docking station is less than a first set distance, the status data of the unmanned surface vessel and the docking station are acquired respectively. The status data includes coordinates, attitude, acceleration and angular velocity. The state data of the unmanned surface vessel and the docking station, as well as the predicted state data of the unmanned surface vessel and the docking station, are fused to generate the target optimal state data of the unmanned surface vessel and the docking station. The predicted state data is obtained based on the historical state data of the unmanned surface vessel and the docking station. Based on the target optimal state data of the unmanned surface vessel and the docking station, a gravitational field with the front end of the docking station as the gravitational source and a repulsive field with the rear end of the docking station and obstacles as the repulsive source are constructed. Calculate the relative positions of the unmanned surface vessel and the docking station based on the target optimal state data of the unmanned surface vessel and the docking station; Based on the relative positions of the unmanned surface vessel and the docking station, the control mode of the unmanned surface vessel is determined, and a virtual force field of the unmanned surface vessel is generated based on the gravitational field, repulsive field, control mode and its corresponding algorithm. Based on the direction and magnitude of the virtual force field, the motion path of the unmanned surface vessel is determined, and the unmanned surface vessel is controlled to dock with the docking station according to the motion path of the unmanned surface vessel.

2. The method according to claim 1, characterized in that, Predicted state data is obtained based on historical state data of unmanned surface vessels and docking stations, including: Based on the angular velocity and acceleration shown by the historical state data of the previous set time, the intermediate predicted state data of the current time is predicted by integrating the angular velocity of the inertial measurement unit. The intermediate predicted state data and historical state data are fused to obtain the predicted state data for the current moment.

3. The method according to claim 2, characterized in that, The system integrates the status data of the unmanned surface vessel and the docking station, as well as the predicted status data of the unmanned surface vessel and the docking station, including: Obtain the state error covariance corresponding to the predicted state data of the unmanned surface vessel and the docking station; The global coordinates in the status data of the unmanned surface vessel and docking station are converted into Cartesian coordinates in the local Cartesian coordinate system. The global coordinates include longitude, latitude and elevation. The heading angle shown by the converted Cartesian coordinates and state data is used as the actual observation vector, and the observation equation is constructed based on the actual observation vector to obtain the deviation. The deviation is used to show the degree of deviation between the predicted state data and the actual observation vector. Calculate the Kalman gain based on the state error covariance, the observation matrix, and the preset measurement noise covariance; Based on the Kalman gain, the bias, and the predicted state data of the unmanned surface vessel and the docking station, the optimal state data of the target is calculated, and the optimal state data of the target is used as the predicted state data for the next set time.

4. The method according to claim 1, characterized in that, The repulsive field includes a first repulsive field and a second repulsive field; correspondingly Based on the target optimal state data of the unmanned surface vessel and the docking station, a gravitational field with the front end of the docking station as the gravitational source and a repulsive field with the rear end of the docking station and obstacles as repulsive sources are constructed, including: Using the coordinates of the docking station's front end, as shown by the target optimal state data of the unmanned surface vessel and docking station, as the center of the gravity source, a gravity field is constructed. This gravity field is used to determine gravity and its direction based on the gravity function. Using the coordinates of the rear end of the docking station shown in the target optimal state data of the unmanned surface vessel and the docking station as the center of the repulsive force source, a first repulsive force field is constructed. The first repulsive force field is used to determine the first repulsive force and the direction of the first repulsive force according to the first repulsive force function. The coordinates of all obstacles on the unmanned surface vessel are obtained, and a second repulsive field is constructed based on the coordinates of all obstacles. The second repulsive field is used to determine the second repulsive force and the direction of the repulsive force based on the second repulsive force function.

5. The method according to claim 4, characterized in that, The relative position includes the relative distance, relative azimuth, and relative heading deviation between the unmanned surface vessel and the docking station; based on the relative position between the unmanned surface vessel and the docking station, the control mode of the unmanned surface vessel is determined, including: If the relative distance between the unmanned surface vessel and the docking station meets the preset contact range and the relative heading deviation between the unmanned surface vessel and the docking station is less than the set deviation range, the control mode is determined to be the forward alignment mode. If the relative azimuth angle between the unmanned surface vessel and the docking station is greater than the set azimuth angle and the relative distance between the unmanned surface vessel and the docking station is less than a set multiple of the length of the docking station, the control mode is determined to be the rearward bypass mode. If the relative heading deviation between the unmanned surface vessel and the docking station is greater than the set deviation range, the control mode of the unmanned surface vessel is determined to be the alignment mode.

6. The method according to claim 5, characterized in that, The virtual force field of the unmanned surface vessel is generated based on the gravitational field, repulsive field, control mode, and corresponding algorithm, including: In the forward alignment control mode, a gravitational vector is calculated based on the gravitational function of the gravitational field, a first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and a second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the weight of the gravitational vector using a gravitational line alignment algorithm, it is superimposed with the first and second repulsive vectors to obtain a resultant force vector. A virtual force field is constructed based on the resultant force vector. In the control mode of rearward bypass mode, a gravitational vector is calculated based on the gravitational function of the gravitational field, a first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and a second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the weight of the first repulsive vector using a repulsive point guidance algorithm, it is superimposed with the gravitational vector and the second repulsive vector to obtain a resultant force vector. A virtual force field is constructed based on the resultant force vector. In the alignment control mode, a gravitational vector is calculated based on the gravitational function of the gravitational field, a first repulsive vector is calculated based on the first repulsive function of the first repulsive field, and a second repulsive vector is calculated based on the second repulsive function of the second repulsive field. After adjusting the lateral weight and forward weight of the gravitational vector using the gravitational line alignment algorithm, it is superimposed with the first repulsive vector and the second repulsive vector to obtain a resultant force vector. A virtual force field is constructed based on the resultant force vector.

7. The method according to claim 1, characterized in that, The motion path of the unmanned surface vessel is determined based on the direction and magnitude of the virtual force field, including: The path point generation rules are determined based on the direction and magnitude of the virtual force field, and the generation rules include the spacing and direction of the path points. Within each control cycle, the next path point is generated based on the corresponding generation rules and the direction and magnitude of the virtual force field; By connecting a series of path points in sequence, the motion path of the unmanned surface vessel is generated.

8. The method according to claim 7, characterized in that, Controlling the unmanned surface vessel to dock with the docking station based on the motion path of the unmanned vessel includes: Based on the motion path, calculate the required acceleration and angular velocity of the unmanned surface vessel at each point on the path, and generate corresponding motion commands. The motion command is decomposed into thruster differential speed and rudder angle control quantities, and anti-saturation optimization is performed in combination with the unmanned surface vessel dynamics model; The optimized motion commands are sent to the unmanned surface vessel, driving the unmanned vessel's motors and servos to perform the docking action with the docking station.

9. A docking system for an unmanned surface vessel and a docking station, characterized in that, The system includes: The unmanned surface vessel is equipped with a dual-antenna system and an inertial measurement unit to acquire its own status data. The docking station is equipped with a dual-antenna system and an inertial measurement unit for acquiring its own status data; the control system is deployed on the unmanned surface vessel and includes a processor and a memory communicatively connected to the processor. The memory stores instructions that can be executed by the processor, and the instructions are executed by the processor to enable the processor to perform the method of any one of claims 1-8. An inter-device communication system is used to control data communication between the unmanned surface vessel and the docking station through a communication server and encryption algorithms; A remote monitoring communication system for monitoring data from the communication server and the front end via a communication server and encryption algorithms.

10. The system according to claim 9, characterized in that, The unmanned surface vessel is also equipped with a lidar, which is used to detect obstacles during the docking process and to trigger an emergency stop when the distance to the docking station is less than a second set distance. The unmanned surface vessel is also used for dynamic interception via a command verification module. The unmanned surface vessel performs dynamic interception through a command verification module, including: when the detected speed is greater than a first set speed or the angular velocity is greater than a first set angular velocity, the speed is limited through the command over-limit interception function of the command verification module.

Images