An autonomous swimming method and system for an underwater vehicle based on motion capture

Abstract

The application provides a path planning method for autonomous swimming of an underwater vehicle, wherein a reflective ball is arranged on an obstacle in water and on the underwater vehicle, a lower computer is arranged in the underwater vehicle, a plurality of high-speed cameras are arranged underwater, the high-speed cameras are subjected to ambient light denoising treatment, the high-speed cameras are subjected to spatial position calibration and dynamic calibration in sequence, position information and attitude information are converted into three-dimensional point data and a three-dimensional model in sequence by the upper computer, the heading information of the underwater vehicle is calculated by the upper computer according to the three-dimensional model, and the heading information is transmitted to the lower computer, so that the underwater vehicle avoids the obstacle according to the heading information; the application can be completely applied to a small range of water, and the computing power requirement for the underwater vehicle is smaller.

Description

Technical Field

[0001] This invention relates to the field of attitude control technology for underwater vehicles, and more specifically, to an autonomous swimming method and system for underwater vehicles based on motion capture. Background Technology

[0002] In fields such as marine ranching monitoring and fisheries management, electronic fences are often used to delineate different water areas. Within the water area enclosed by the electronic fence, specialized underwater vehicles, such as manta ray-inspired underwater vehicles, can detect water quality at different depths and in different areas within the electronic fence, helping to monitor the health of marine life. At the same time, underwater vehicles can also use cameras and sensors to check the integrity, stability, and functionality of underwater aquaculture cages, planting facilities, and other structures.

[0003] Within the water area enclosed by an electronic fence, various obstacles such as buoys and aquaculture cages are typically placed. In this field, one type of underwater vehicle is controlled manually, requiring human intervention. When encountering deep waters with many obstacles, manual control becomes difficult, easily leading to collisions and damage to the underwater vehicle. Another traditional type of underwater vehicle calculates its movement path in real time based on the location of obstacles using a series of autonomous underwater swimming algorithms. While this method is highly effective, it requires the installation of various sensors on the underwater vehicle, thus placing high demands on its computing power, specifically the computing power of its built-in processor. Summary of the Invention

[0004] The problem solved by this invention is to provide an autonomous swimming method and system for underwater vehicles based on motion capture, which can be fully applied to small-scale water areas such as marine ranching monitoring and fisheries management, and requires less computing power from the underwater vehicle.

[0005] To address the above problems, this invention provides a path planning method for autonomous underwater vehicles, comprising the following steps:

[0006] S1. Each obstacle in the water and the underwater vehicle is equipped with a reflective ball to provide feedback on the position and attitude information of the obstacle and the underwater vehicle. The underwater vehicle is equipped with a lower-level computer to control the course of the underwater vehicle. Multiple high-speed cameras are installed underwater. The high-speed cameras are used to capture the position and attitude information of the underwater vehicle and the obstacle based on the reflected light from the reflective balls, and transmit the position and attitude information to the upper-level computer.

[0007] S2. Perform ambient light noise reduction processing on the high-speed camera;

[0008] S3. Perform spatial and dynamic calibration on the high-speed camera in sequence to correct the geodetic coordinate system;

[0009] S4. Obtain position and attitude information through a high-speed camera, and convert the position and attitude information into three-dimensional point data of the underwater vehicle and obstacles through a host computer;

[0010] S5. Convert the 3D point data into 3D models of underwater vehicles and obstacles using a host computer.

[0011] S6. The host computer calculates the heading information of the underwater vehicle based on the three-dimensional model and transmits the heading information to the slave computer, so that the underwater vehicle can avoid obstacles according to the heading information.

[0012] The beneficial effects of this method are that it can capture the position and attitude information of obstacles and underwater vehicles by using a high-speed camera based on the reflected light from the reflective ball. Then, the host computer calculates the heading information of the underwater vehicle and transmits the heading information to the slave computer, enabling the slave computer to control the underwater vehicle to avoid obstacles. This method requires less computing power from the underwater vehicle.

[0013] Furthermore, there are multiple reflective spheres on the underwater vehicle, with at least one reflective sphere disposed on the outer surface of the underwater vehicle at its center of mass; there are multiple reflective spheres on each obstacle, and at least one reflective sphere is disposed on the outer surface of the center of mass of each obstacle.

[0014] This setting can more accurately reflect the position and attitude information of obstacles and underwater vehicles. When the center of mass is inside the obstacle or underwater vehicle, the reflective ball can be placed on the outer surface of the center of mass of the obstacle or underwater vehicle without affecting the high-speed camera's capture. Only the corresponding compensation value needs to be set in the host computer.

[0015] Furthermore, the lower-level machine includes an MCU, and the CPG phase oscillator model in the MCU is as follows:

[0016]

[0017] The first equation is the amplitude equation, a i Indicates amplitude. The second derivative represents the amplitude. The first derivative of the amplitude, γ i A represents the positive constant controlling the convergence rate of the amplitude. i The first equation represents the desired amplitude; the second equation is the phase equation, φ. i φ represents the phase of the i-th unit. j This represents the phase of the j-th unit. f represents the first derivative of the phase of the i-th unit. i Represents the natural frequency, ω ij This represents the coupling weight of the j-th unit to the i-th unit. The third equation represents the desired phase difference; the third equation is the output equation, θ. i Indicates the output value; where a i γ i φ i ω ij A is a state parameter. i f i , These are the input control parameters.

[0018] This model enables the rhythmic signals generated by the CPG phase oscillator to effectively control the course of the underwater vehicle.

[0019] Furthermore, in step S2, the process of eliminating ambient light noise from the high-speed camera specifically includes the following steps:

[0020] S21. To block strong ambient light in the water, an L-shaped pole is designated as L-pole. At least four reflective spheres are installed on the L-pole, with two reflective spheres installed on the long axis of the L-pole, one reflective sphere installed at the vertical connection point of the L-pole, and the other reflective sphere installed on the short axis of the L-pole.

[0021] S22, Preset the range of reflected light intensity of the reflective sphere that the high-speed camera can capture;

[0022] S23. Place the L-rod horizontally on the bottom of the water;

[0023] S23. Capture the reflective spheres on the L-bar using all high-speed cameras, and adjust the intensity range of the reflected light until all reflective spheres can be captured by all high-speed cameras.

[0024] This method allows the high-speed camera to preprocess the image, thus preventing the high-speed camera from failing to capture the reflective sphere in subsequent processes.

[0025] Furthermore, the spatial location calibration in step S3 specifically includes the following steps:

[0026] S31. Capture the reflective ball on the L-shaped rod lying flat on the bottom of the water using all high-speed cameras to obtain the position and attitude information of the L-shaped rod;

[0027] S32. Convert the position and attitude information of the L-bar into three-dimensional point data of the L-bar using the host computer;

[0028] S33. Generate a 3D model of the L-shaped rod from the 3D point data using the host computer;

[0029] S34. Based on the 3D model, determine whether the L-rod is formed using the host computer. If it is, mark the major axis of the L-rod as the X-axis, the minor axis of the L-rod as the Y-axis, and the water depth direction as the Z-axis. If not, adjust the exposure time of the high-speed camera and return to step S2.

[0030] This method can correct the geodetic coordinate system and generate a three-dimensional coordinate axis in the host computer that is suitable for calculation and conforms to the actual water conditions.

[0031] Furthermore, the dynamic calibration in step S3 specifically includes the following steps:

[0032] S301. A rod of fixed length is denoted as rod T.

[0033] S302. A reflective ball is installed at each end of the T-bar;

[0034] S303. Submerge the T-pole in the water and move it, and within a fixed time period, capture the reflective ball on the T-pole using all high-speed cameras to obtain the position and attitude information of the T-pole.

[0035] S304. Convert the position and attitude information of the T-bar into three-dimensional point data of the T-bar using the host computer;

[0036] S305. Generate a 3D model of the T-bar from the 3D point data of the T-bar using the host computer;

[0037] S306. Based on the 3D model of the T-bar, the host computer determines whether all high-speed cameras can capture the reflective ball on the T-bar at different time periods. If yes, end; if no, adjust the exposure time of the high-speed cameras and return to step S2.

[0038] This method ensures that high-speed cameras can capture moving reflective spheres, thereby ensuring the capture of moving underwater vehicles and obstacles.

[0039] Further, in step S6, calculating the heading information of the underwater vehicle specifically includes the following steps: S61, calculating the resultant force function F(q); wherein, the potential field function U of the gravitational potential field... att (q) is as follows;

[0040]

[0041] In the above formula, η is the proportional gain coefficient; ρ(q,q g () is a vector, with a magnitude equal to the underwater vehicle's position q and the target point q. g Euclidean distance between | qq g |, the direction is the direction in which the underwater vehicle points to the target point;

[0042] The negative gradient force function F of the gravitational potential fieldatt (q) is as follows:

[0043]

[0044] Potential function U of repulsive potential field rep (q) is as follows:

[0045]

[0046] In the above two formulas, Represents the potential field function U att The negative gradient of (q); k is a positive proportionality coefficient; ρ0 is a constant, representing the maximum range of influence of the obstacle on the underwater vehicle;

[0047] The negative gradient force function F of the repulsive potential field rep (q) is as follows:

[0048]

[0049] The resultant potential field function U(q) is as follows:

[0050] U(q)=U att (q)+U rep (q);

[0051] Then, the resultant force function F(q) is obtained as follows:

[0052]

[0053] S62. On the two-dimensional plane formed by the X-axis and Y-axis, the two vector directions of the resultant force function F(q) are taken as the heading of the underwater vehicle.

[0054] This method can effectively calculate the heading information of an underwater vehicle, enabling the underwater vehicle to avoid obstacles based on the heading.

[0055] Furthermore, the reflective sphere is a passive infrared reflective sphere, and the high-speed camera is an infrared high-speed camera.

[0056] Infrared light has a longer wavelength than visible light, which allows the reflective sphere to be captured more effectively.

[0057] The present invention also provides an autonomous swimming system for an underwater vehicle based on motion capture, comprising:

[0058] Multiple obstacle modules, each including a reflective sphere unit for feeding back the position and attitude information of the obstacle module;

[0059] The underwater vehicle module includes a reflective ball unit for feeding back the position and attitude information of the underwater vehicle module, and a lower-level computer unit for controlling the underwater vehicle module.

[0060] Multiple high-speed camera modules, each used to capture the position and attitude information of the obstacle module and the underwater vehicle module based on the reflective ball unit;

[0061] The water module contains all obstacle modules, high-speed camera modules, and underwater vehicle modules.

[0062] The host computer module communicates with the slave computer unit and the high-speed camera module. The host computer module is used to display and process the position and attitude information captured by the high-speed camera module, and transmit the processed position and attitude information to the slave computer unit, so that the slave computer unit can control the underwater vehicle module according to the received position and attitude information. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the method steps in this invention;

[0064] Figure 2 This is a schematic diagram of the system in practical application of the present invention;

[0065] Figure 3 This is a schematic diagram of a manta ray-inspired underwater vehicle with reflective spheres in an embodiment of the present invention;

[0066] Figure 4 This is a schematic diagram of a three-dimensional model formed from three-dimensional point data of a manta ray-inspired underwater vehicle in an embodiment of the present invention; Detailed Implementation

[0067] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0068] This embodiment describes a path planning method for autonomous underwater vehicles, including the following steps:

[0069] S1. Each obstacle in the water and the underwater vehicle is equipped with a reflective ball to provide feedback on the position and attitude information of the obstacle and the underwater vehicle. The underwater vehicle is equipped with a lower-level computer to control the course of the underwater vehicle. Multiple high-speed cameras are installed underwater. The high-speed cameras are used to capture the position and attitude information of the underwater vehicle and the obstacle based on the reflected light from the reflective balls, and transmit the position and attitude information to the upper-level computer.

[0070] S2. Perform ambient light noise reduction processing on the high-speed camera;

[0071] S3. Perform spatial and dynamic calibration on the high-speed camera in sequence to correct the geodetic coordinate system;

[0072] S4. Obtain position and attitude information through a high-speed camera, and convert the position and attitude information into three-dimensional point data of the underwater vehicle and obstacles through a host computer;

[0073] S5. Convert the 3D point data into 3D models of underwater vehicles and obstacles using a host computer.

[0074] S6. The host computer calculates the heading information of the underwater vehicle based on the three-dimensional model and transmits the heading information to the slave computer, so that the underwater vehicle can avoid obstacles according to the heading information.

[0075] Furthermore, there are multiple reflective spheres on the underwater vehicle, with at least one reflective sphere located on the outer surface of the underwater vehicle's center of mass; there are multiple reflective spheres on each obstacle, and at least one reflective sphere is located on the outer surface of each obstacle's center of mass.

[0076] Furthermore, the lower-level machine includes the MCU, and the CPG phase oscillator model in the MCU is as follows:

[0077]

[0078] The first equation is the amplitude equation, a i Indicates amplitude. The second derivative represents the amplitude. The first derivative of the amplitude, γ i A represents the positive constant controlling the convergence rate of the amplitude. i The first equation represents the desired amplitude; the second equation is the phase equation, φ. i φ represents the phase of the i-th unit. j This represents the phase of the j-th unit. f represents the first derivative of the phase of the i-th unit. i Represents the natural frequency, ω ij This represents the coupling weight of the j-th unit to the i-th unit. The third equation represents the desired phase difference; the third equation is the output equation, θ. i Indicates the output value; where a i γ i φ i ω ij A is a state parameter. i f i , These are the input control parameters.

[0079] Furthermore, the specific steps for eliminating ambient light noise in high-speed cameras include:

[0080] S21. To block strong ambient light in the water, an L-shaped pole is designated as L-pole. At least four reflective spheres are installed on the L-pole, with two reflective spheres installed on the long axis of the L-pole, one reflective sphere installed at the vertical connection point of the L-pole, and the other reflective sphere installed on the short axis of the L-pole.

[0081] S22, Preset the range of reflected light intensity of the reflective sphere that the high-speed camera can capture;

[0082] S23. Place the L-rod horizontally on the bottom of the water;

[0083] S23. Capture the reflective spheres on the L-bar using all high-speed cameras, and adjust the intensity range of the reflected light until all reflective spheres can be captured by all high-speed cameras.

[0084] Furthermore, the spatial location calibration in step S3 specifically includes the following steps:

[0085] S31. Capture the reflective ball on the L-shaped rod lying flat on the bottom of the water using all high-speed cameras to obtain the position and attitude information of the L-shaped rod;

[0086] S32. Convert the position and attitude information of the L-bar into three-dimensional point data of the L-bar using the host computer; S33. Generate a three-dimensional model of the L-bar from the three-dimensional point data using the host computer;

[0087] S34. Based on the 3D model, determine whether the L-rod is formed using the host computer. If it is, mark the major axis of the L-rod as the X-axis, the minor axis of the L-rod as the Y-axis, and the water depth direction as the Z-axis. If not, adjust the exposure time of the high-speed camera and return to step S2.

[0088] Furthermore, the dynamic calibration in step S3 specifically includes the following steps:

[0089] S301. A rod of fixed length is denoted as rod T.

[0090] S302. A reflective ball is installed at each end of the T-bar;

[0091] S303. Submerge the T-pole in the water and move it, and within a fixed time period, capture the reflective ball on the T-pole using all high-speed cameras to obtain the position and attitude information of the T-pole.

[0092] S304. Convert the position and attitude information of the T-bar into three-dimensional point data of the T-bar using the host computer;

[0093] S305. Generate a 3D model of the T-bar from the 3D point data of the T-bar using the host computer;

[0094] S306. Based on the 3D model of the T-bar, the host computer determines whether all high-speed cameras can capture the reflective ball on the T-bar at different time periods. If yes, end; if no, adjust the exposure time of the high-speed cameras and return to step S2.

[0095] Furthermore, in step S6, calculating the heading of the underwater vehicle specifically includes the following steps: S61, calculating the resultant force function F(q); where the gravitational potential field function U... att (q) is as follows;

[0096]

[0097] In the above formula, η is the proportional gain coefficient; ρ(q,q g () is a vector, with a magnitude equal to the underwater vehicle's position q and the target point q. g Euclidean distance between | qq g |, the direction is the direction in which the underwater vehicle points to the target point;

[0098] The negative gradient force function F of the gravitational potential field att (q) is as follows:

[0099]

[0100] Potential function U of repulsive potential field rep (q) is as follows:

[0101]

[0102] In the above two formulas, Represents the potential field function U att The negative gradient of (q); k is a positive proportionality coefficient; ρ0 is a constant, representing the maximum range of influence of the obstacle on the underwater vehicle;

[0103] The negative gradient force function F of the repulsive potential field rep (q) is as follows:

[0104]

[0105] The resultant potential field function U(q) is as follows:

[0106] U(q)=U att (q)+U rep (q);

[0107] Then, the resultant force function F(q) is obtained as follows:

[0108]

[0109] S62. On the two-dimensional plane formed by the X-axis and Y-axis, the two vector directions of the resultant force function F(q) are taken as the heading of the underwater vehicle.

[0110] Specifically, the obstacles mainly consist of buoys and aquaculture cages within a fixed water area, primarily aquaculture cages. Fish and other aquatic organisms are not considered obstacles because the underwater vehicle's speed is relatively slow, and the fish and other organisms have the self-awareness to avoid obstacles. Therefore, fish and other aquatic organisms are not considered obstacles and do not require reflective spheres. The number of reflective spheres on the obstacles can be flexibly adjusted according to their actual size. Each obstacle has at least one reflective sphere fixed on its outer surface at the center of mass. The reflective sphere on the outer surface at the center of mass can fully reflect the obstacle's position information. The remaining reflective spheres on each obstacle are fixedly placed at the edges to present the obstacle's edge outline, thereby displaying attitude information. The same applies to the underwater vehicle. In this embodiment, as shown in the attached... Figure 3 As shown, the underwater vehicle is a manta ray-inspired underwater vehicle, with at least one reflective sphere fixed to the outer surface at the center of mass of the manta ray-inspired underwater vehicle. The remaining reflective spheres are fixed to the wings and head of the manta ray-inspired underwater vehicle. This allows all the reflective spheres on the manta ray-inspired underwater vehicle to form an image in the host computer after being captured by a high-speed camera, as shown in the attached image. Figure 4 The 3D model shown has reflective spheres on the outer surface of the center of mass that can fully reflect the position information of obstacles, and reflective spheres on the wings and head that can present the edge outline of the manta ray-like underwater vehicle, thus displaying attitude information; the MCU is STM32.

[0111] In this embodiment, the host computer is a computer device. The host computer uses motion capture software to convert the position and attitude information of the underwater vehicle and obstacles into 3D point data and a 3D model. The motion capture software is Qualisys Track Manager. High-speed cameras need to be evenly distributed around the water area. In step S21, strong ambient light in the water is blocked using mosaic processing in the host computer software. Strong ambient light refers to light that cannot be eliminated by adjusting the parameters of each high-speed camera, so it needs to be blocked using mosaic. The spatial position calibration in step S3 is shown in the attached figure. Figure 4As shown, when the L-shaped rod is laid flat, if the shape of the L-shaped rod is observed to be formed in the host computer, that is, both the major and minor axes are clearly displayed, then the major axis of the L-shaped rod is marked as the X-axis, the minor axis of the L-shaped rod is marked as the Y-axis, and the water depth direction is marked as the Z-axis. In this way, a three-dimensional coordinate axis that conforms to the actual water conditions can be formed in the host computer for calculation, thereby completing the spatial position calibration. Dynamic calibration is to ensure that the high-speed camera can capture the moving reflective ball, and thus be able to capture the moving underwater vehicle and obstacles. The underwater vehicle's heading is only on the XY plane, and there is also a Z-axis, which can be changed by the vertical offset of the tail fin. In fact, it is to convert the three-dimensional motion into a two-dimensional plane plus the Z-axis. The tail fin and pectoral fin are actually independent of each other. In this embodiment, the heading is changed on the two-dimensional plane by the action of the underwater vehicle's pectoral fin.

[0112] In addition, in this embodiment, the three-dimensional coordinates of each reflective sphere on the manta ray-inspired underwater vehicle and the three-dimensional coordinates of each reflective sphere on the obstacle are displayed on the host computer using MATLAB, and then the heading information is calculated through step S6.

[0113] In a preferred embodiment of the invention, the reflective sphere is a passive infrared reflective sphere, and the high-speed camera is an infrared high-speed camera. Infrared light has a longer wavelength than conventional light, making it more advantageous for capturing images.

[0114] As attached Figure 2 As shown, the present invention also provides an autonomous swimming system for underwater vehicles based on motion capture, comprising:

[0115] Multiple obstacle modules, each including a reflective sphere unit for feeding back the position and attitude information of the obstacle module;

[0116] The underwater vehicle module includes a reflective ball unit for feeding back the position and attitude information of the underwater vehicle module, and a lower-level computer unit for controlling the underwater vehicle module.

[0117] Multiple high-speed camera modules, each used to capture the position and attitude information of the obstacle module and the underwater vehicle module based on the reflective ball unit;

[0118] The water module contains all obstacle modules, high-speed camera modules, and underwater vehicle modules.

[0119] The host computer module communicates with the slave computer unit and the high-speed camera module. The host computer module is used to display and process the position and attitude information captured by the high-speed camera module, and transmit the processed position and attitude information to the slave computer unit, so that the slave computer unit can control the underwater vehicle module according to the received position and attitude information.

[0120] While the disclosure is as stated above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of this disclosure, and all such changes and modifications will fall within the protection scope of this invention.

Claims

1. A method for autonomous swimming of an underwater vehicle based on motion capture, characterized in that, Includes the following steps: S1. Each obstacle in the water and the underwater vehicle are equipped with a reflective ball for feeding back the position and attitude information of the obstacle and the underwater vehicle. The underwater vehicle is equipped with a lower-level computer for controlling the course of the underwater vehicle. Multiple high-speed cameras are installed underwater. The high-speed cameras are used to capture the position and attitude information of the underwater vehicle and the obstacle based on the reflected light from the reflective balls, and transmit the position and attitude information to the upper-level computer. S2. Perform ambient light noise reduction processing on the high-speed camera; S3. Perform spatial position calibration and dynamic calibration on the high-speed camera in sequence to correct the geodetic coordinate system; S4. Obtain the position information and attitude information through the high-speed camera, and convert the position information and attitude information into three-dimensional point data of the underwater vehicle and obstacles through the host computer; S5. The host computer converts the three-dimensional point data into a three-dimensional model of the underwater vehicle and obstacles. S6. The host computer calculates the heading information of the underwater vehicle based on the three-dimensional model, and transmits the heading information to the slave computer, so that the underwater vehicle avoids the obstacle according to the heading information. The underwater vehicle has multiple reflective spheres, with at least one reflective sphere disposed on the outer surface of the underwater vehicle at its center of mass; each obstacle has multiple reflective spheres, and at least one reflective sphere is disposed on the outer surface of the center of mass of each obstacle. Step S2, specifically the process of eliminating ambient light noise in the high-speed camera, includes the following steps: S21. To block strong ambient light in the water, an L-shaped rod is referred to as the L rod. At least four reflective balls are set on the L rod, two of which are set on the long axis of the L rod, one of which is set at the vertical connection of the L rod, and the other of which is set on the short axis of the L rod. S22. Preset the range of reflected light intensity of the reflective sphere that the high-speed camera can capture; S23. Place the L-rod horizontally on the bottom of the water; S23. Capture the reflective spheres on the L-bar using all the high-speed cameras, and adjust the intensity range of the reflected light until all the reflective spheres can be captured by all the high-speed cameras; In step S3, the host computer determines whether the L-rod is formed; if so, the major axis of the L-rod is marked as the X-axis, the minor axis of the L-rod is marked as the Y-axis, and the water depth direction is marked as the Z-axis. The dynamic calibration process specifically includes the following steps: S301. A rod of fixed length is denoted as rod T. S302. A reflective ball is installed at each end of the T-bar; S303. Submerge the T-bar in the water and move it, and within a fixed time period, capture the reflective ball on the T-bar through all the high-speed cameras to obtain the position and attitude information of the T-bar. S304. The position and attitude information of the T-bar are converted into three-dimensional point data of the T-bar by the host computer; S305. Generate a three-dimensional model of the T-bar from the three-dimensional point data of the T-bar using a host computer; S306. Based on the three-dimensional model of the T-bar, the host computer determines whether all the high-speed cameras can capture the reflective ball on the T-bar in different time periods. If yes, end; if no, adjust the exposure time of the high-speed cameras and return to step S2. Step S6, calculating the heading information of the underwater vehicle specifically includes the following steps: S61. Calculate the resultant force function Among them, the potential field function of the gravitational potential field. as follows: ； In the above formula, Proportional gain coefficient; It is a vector, and its magnitude is the position of the underwater vehicle. and target point Euclidean distance between The direction is the direction in which the underwater vehicle points to the target point; Negative gradient force function of gravitational potential field as follows: ； Potential function of repulsive potential field as follows: ； In the formula, Potential field function The negative gradient; It is a positive proportionality coefficient; is a constant, representing the maximum range of influence of the obstacle on the underwater vehicle; Negative gradient force function of repulsive potential field as follows: ； Resultant potential field function as follows: ； Then, the resultant force function is obtained. as follows: ； S62. On the two-dimensional plane formed by the X-axis and Y-axis, the resultant force function The two vector directions are used as the heading of the underwater vehicle.

2. The autonomous swimming method for an underwater vehicle based on motion capture according to claim 1, characterized in that, The lower-level machine includes an MCU, and the CPG phase oscillator model in the MCU is as follows: ； The first equation is the amplitude equation. Indicates amplitude, The second derivative represents the amplitude. The first derivative represents the amplitude. This represents a positive constant that controls the convergence rate of the amplitude. The first equation represents the desired amplitude; the second equation is the phase equation. This represents the phase of the i-th unit. This represents the phase of the j-th unit. The first derivative of the phase of the i-th unit is represented by . Indicates the natural frequency. This represents the coupling weight of the j-th unit to the i-th unit. The third equation represents the desired phase difference; the fourth equation is the output equation. Indicates the output value; where , , , For state parameters, , , Input control parameters.

3. The autonomous swimming method for an underwater vehicle based on motion capture according to claim 1, characterized in that, Step S3, spatial location calibration, specifically includes the following steps: S31. Capture the reflective ball on the L-bar lying flat on the bottom of the water using all the high-speed cameras to obtain the position and attitude information of the L-bar; S32. The position and attitude information of the L-rod are converted into three-dimensional point data of the L-rod by the host computer; S33. Generate a three-dimensional model of the L-shaped rod from the three-dimensional point data using a host computer; S34. Based on the three-dimensional model, the host computer determines whether the L-rod is formed. If so, the major axis of the L-rod is marked as the X-axis, the minor axis of the L-rod is marked as the Y-axis, and the depth direction of the water is marked as the Z-axis. If not, the exposure time of the high-speed camera is adjusted, and the process returns to step S2.

4. The autonomous swimming method for an underwater vehicle based on motion capture according to claim 1 or 3, characterized in that, The reflective sphere is a passive infrared reflective sphere, and the high-speed camera is an infrared high-speed camera.

5. An autonomous swimming system for an underwater vehicle based on motion capture, characterized in that, The autonomous swimming method for an underwater vehicle based on motion capture as described in any one of claims 1-4 is characterized by comprising: Multiple obstacle modules, each of which includes a reflective ball unit for feeding back the position and attitude information of the obstacle module; An underwater vehicle module, comprising a reflective ball unit for feeding back the position and attitude information of the underwater vehicle module, and a lower-level computer unit for controlling the underwater vehicle module; Multiple high-speed camera modules, each of which is used to capture the position and attitude information of the obstacle module and the underwater vehicle module based on the reflective ball unit; The water area module, in which all the aforementioned obstacle modules, high-speed camera modules and underwater vehicle modules are housed, is located; The host computer module is communicatively connected to the slave computer unit and the high-speed camera module. The host computer module is used to display and process the position information and attitude information captured by the high-speed camera module, and transmit the processed position information and attitude information to the slave computer unit, so that the slave computer unit controls the underwater vehicle module according to the received position information and attitude information.

Images