A narrow space-oriented machine manta ray vision servo autonomous precise benthic control method

Abstract

The present application relates to underwater vehicle control technology field, specifically to a kind of machine manta ray visual servo autonomous precision benthic control method and system for narrow space, comprising: setting the initialization parameter of machine manta ray fixed-point benthic;Obtain the relative pose information between machine manta ray and the mark between fixed-point benthic position;Obtain the real-time phase difference corresponding to each pectoral fin of machine manta ray;Obtain the buoyancy of machine manta ray;Machine manta ray is controlled benthically.The present application can not only realize trajectory convergence and position convergence, but also can guarantee attitude stable convergence, embodies the good convergence and operating stability of the present application in actual operation.

Description

Technical Field

[0001] This invention relates to the field of underwater vehicle control technology, specifically to a visual servo autonomous and precise benthic control method and system for robotic manta rays operating in confined spaces. Background Technology

[0002] Manta rays are a type of fish that use their pectoral fins as their main propulsion and posture control organs. They move by flapping their pectoral fins with large amplitude and low frequency to generate continuous thrust and lift in the water, enabling them to move forward, ascend, and dive.

[0003] The robotic manta ray is developed based on the biomimetic manta ray. As a highly efficient biomimetic submersible, the robotic manta ray achieves smooth and stable maneuvering in various aquatic environments through the flapping of its pectoral fins and the adjustment of its tail fin. For tasks such as long-term fixed-point observation or bottom sediment sampling, the robotic manta ray often needs to switch to a landing mode, using seabed support to achieve a stable stay for a long time.

[0004] The key to autonomous benthic control for robotic manta rays lies in their depth control. Existing autonomous benthic control methods are mainly based on two principles: (1) adjusting buoyancy by changing the volume of water to achieve depth changes; (2) adjusting the pitch angle of the tail fin to achieve diving when forward velocity is available. Based on the above principles, there are currently three main technical solutions: (1) using a buoyancy adjustment mechanism to obtain vertical velocity to achieve benthic control; (2) adjusting the pitch angle by adjusting the tail fin amplitude offset to achieve benthic control; (3) combining tail fin offset with a buoyancy adjustment mechanism to achieve benthic control.

[0005] Bottom-dwelling methods employing buoyancy adjustment mechanisms achieve heave and sag by altering buoyancy through changes in displacement volume. While offering advantages such as high control stability and smooth movement, they are prone to horizontal drift under water flow disturbances and lack effective constraints on horizontal position, making it difficult to achieve fixed-point vertical bottom-dwelling. Bottom-dwelling methods based on tail fin pitch control, when possessing a certain forward velocity, generate downward lift by adjusting the tail fin offset. This method is structurally simple and the movement is continuous and smooth, but it is highly dependent on forward velocity and easily introduces horizontal displacement errors under water flow disturbances. Furthermore, it requires a certain forward distance, making it difficult to meet the needs of operations in confined spaces. Composite bottom-dwelling methods, by adjusting the amplitude and offset of the tail fin and combining it with a buoyancy mechanism, allow the robotic manta ray to dive in different postures, incorporating advantages such as low-impact bottom-dwelling. Examples include existing bottom-dwelling control methods for manta ray-inspired vehicles based on tail fins and variable buoyancy systems. However, this patent still relies on forward movement and is limited by the forward movement distance, failing to mention precise bottom-dwelling control strategies in confined spaces. It is evident that existing autonomous benthic control methods have specific advantages in various aspects, but they generally suffer from insufficient resistance to water disturbance, limited accuracy of benthic sites, and high requirements for operational space scale.

[0006] Therefore, there is a need to provide a visual servo autonomous and precise benthic control method for robotic manta rays in confined spaces to solve the above problems. Summary of the Invention

[0007] While existing autonomous benthic control methods have specific advantages in various aspects, they generally suffer from insufficient resistance to water disturbance, limited benthic point accuracy, and high requirements for operational space scale. This invention provides a visual servo-based autonomous and precise benthic control method for robotic manta rays in confined spaces to solve these problems.

[0008] The first aspect of this invention provides a visual servo-based autonomous and precise benthic control method for robotic manta rays in confined spaces, employing the following technical solution, including: Set the initialization parameters for the robotic manta ray to be stationary on the bottom. The initialization parameters include: the reference amplitude of the two sides of the pectoral fin, the reference phase difference, the flapping frequency, and the preset height for the stationary bottom task. The relative pose information between the robotic manta ray and the marker at the fixed benthic location is obtained. The relative pose information includes the distance components in the X-axis direction, Y-axis direction, and Z-axis direction, as well as the relative yaw angle between the robotic manta ray and the marker at the fixed benthic location. Based on the preset height of the fixed-point benthic task, the distance component in the Z-axis direction, the height of the work area, the top radius of the work area, and the landing area radius of the work area, the radius of the dynamic safety threshold circle is obtained. The relative distance between the fixed-point benthic position and the robotic manta ray is obtained based on the distance components in the X-axis and Y-axis directions. The yaw angle difference of the robotic manta ray at different horizontal positions is obtained based on the relative distance, relative yaw angle, and dynamic safety threshold circle radius, and the yaw angle change is also obtained. The flapping control parameters of the robotic manta ray are obtained based on the yaw angle difference, the yaw angle change, and a fuzzy PID control algorithm based on the phase difference. The real-time phase difference corresponding to each pectoral fin of the robotic manta ray is obtained based on the flapping control parameters. The real-time altitude of the robotic manta ray is assigned as the distance component in the Z-axis direction. The difference between the real-time altitude and the preset altitude of the fixed-point benthic task is used as the acquired altitude difference value, and the difference between the real-time altitude and the altitude at the previous moment is used as the altitude change value. Based on the altitude change value, altitude difference value, and buoyancy, a fuzzy PID control algorithm is used to obtain the buoyancy control parameters of the robotic manta ray. The buoyancy of the robotic manta ray is obtained based on the buoyancy control parameters and the preset altitude of the fixed-point benthic task. Benthic control of robotic manta rays is achieved based on their buoyancy and the real-time phase difference corresponding to each pectoral fin.

[0009] A further technical solution of the present invention is that the expression for the radius of the dynamic safety threshold circle is:

[0010] In the formula, Indicates the radius of the dynamic safety threshold circle; Indicates the top radius of the work area; Indicates the radius of the landing area of ​​the work area; Represents the distance component along the Z-axis; Indicates the preset altitude for fixed-point benthic missions; Indicates the height of the work area.

[0011] A further technical solution of the present invention is that the step of obtaining the location of the fixed-point benthic habitat and the relative distance between the robotic manta ray and the distance component in the X-axis direction is as follows:

[0012] In the formula, This indicates the relative distance between the fixed-location benthic site and the robotic manta ray; Represents the distance component along the X-axis; This represents the distance component along the Y-axis.

[0013] A further technical solution of the present invention is to obtain the difference in yaw angle of the robotic manta ray at different horizontal positions based on relative distance, relative yaw angle, and dynamic safety threshold circle radius, and to obtain the change in yaw angle as follows: The desired yaw angle is obtained based on the relative distance and the radius of the dynamic safety threshold circle; The difference between the relative yaw angle and the desired yaw angle is taken as the yaw angle difference. The difference between the relative yaw angle and the yaw angle at the previous moment is taken as the change in yaw angle.

[0014] A further technical solution of the present invention is that the step of obtaining the desired yaw angle based on the relative distance and the size of the dynamic safety threshold circle radius is as follows: The vector angle is obtained based on the relative distance, the distance component in the X-axis direction, and the distance component in the Y-axis direction; Obtain the target angle along the route based on vector angles; When the relative distance is less than or equal to the radius of the dynamic safety threshold circle, the desired yaw angle is 0; when the relative distance is greater than the radius of the dynamic safety threshold circle, the target angle along the way is used as the desired yaw angle.

[0015] A further technical solution of the present invention is that the steps for obtaining the flapping control parameters of the robotic manta ray based on the yaw angle difference, the yaw angle change, and the fuzzy PID control algorithm based on the phase difference are as follows: The yaw angle difference and yaw angle change are fuzzified and input into the fuzzy PID control algorithm based on phase difference, and the initial flapping control parameters of the robotic manta ray are output. When the distance component in the Y-axis direction is less than or equal to zero, the initial flapping control parameters are used as the flapping control parameters for the robotic manta ray. When the distance component in the Y-axis direction is greater than zero, the negative value of the initial flapping control parameter is used as the flapping control parameter for the robotic manta ray.

[0016] A further technical solution of the present invention is that the step of obtaining the real-time phase difference corresponding to each pectoral fin of the robotic manta ray based on flapping control parameters is as follows:

[0017] In the formula, This indicates the real-time phase difference corresponding to the left pectoral fin; This indicates the real-time phase difference corresponding to the right pectoral fin; This indicates the PID parameter representing the phase difference on the left side; This represents the PID parameter indicating the phase difference on the right. This indicates the baseline phase difference of the left pectoral fin; This indicates the baseline phase difference of the right pectoral fin; This indicates the flapping control parameters.

[0018] A further technical solution of the present invention is that, when acquiring the relative pose information between the robotic manta ray and the marker at the fixed benthic location, if the fixed benthic location is lost, a control parameter preservation strategy is executed. The steps for executing the control parameter preservation strategy are as follows: The attenuation coefficient is set according to the preset steady-state value of the time-varying attenuation factor at the maximum safe holding time threshold and the maximum safe holding time threshold; The time-varying decay factor is obtained based on the decay coefficient and the loss duration; The target control law is obtained based on the effective control law of the previous moment, the system preset safety instructions, and the time-varying decay factor. The target control law is then used to control the robotic manta ray.

[0019] A further technical solution of the present invention is that the steps for benthic control of the robotic manta ray based on its buoyancy and the real-time phase difference corresponding to each pectoral fin are as follows: When the real-time altitude is less than or equal to the preset altitude of the fixed-point benthic task, and the relative distance between the fixed-point benthic position and the robotic manta ray is less than or equal to the radius of the landing area of ​​the work area, the benthic control task is successful, and the phase difference of each pectoral fin of the robotic manta ray is controlled to be 0, the flapping frequency is controlled to be 0, and the buoyancy of the robotic manta ray is controlled to be 0. When the real-time altitude is greater than the preset altitude of the fixed-point benthic mission, and the relative distance between the fixed-point benthic position and the robotic manta ray is greater than the radius of the landing area of ​​the operation area, the benthic control mission continues, assigning the current relative yaw angle to the previous relative yaw angle, assigning the current real-time altitude to the previous real-time altitude, and storing the control rate at the current moment.

[0020] A second aspect of the present invention provides a visual servo autonomous precision benthic control system for robotic manta rays operating in confined spaces, the system comprising: The initialization module is used to set the initialization parameters for the robotic manta ray to be stationary on the bottom. The initialization parameters include: the reference amplitude of the two sides of the pectoral fin, the reference phase difference, the flapping frequency, and the preset height for the stationary bottom task. The visual information module is used to acquire the relative pose information between the robotic manta ray and the marker at the fixed-point benthic location. The relative pose information includes the distance components in the X-axis direction, Y-axis direction, and Z-axis direction, as well as the relative yaw angle between the robotic manta ray and the marker at the fixed-point benthic location. Based on the preset height of the fixed-point benthic task, the distance component in the Z-axis direction, the height of the work area, the top radius of the work area, and the landing area radius of the work area, the dynamic safety threshold circle radius is acquired. The phase difference acquisition module is used to obtain the relative distance between the fixed-point benthic position and the robotic manta ray based on the distance components in the X-axis and Y-axis directions. Based on the relative distance, relative yaw angle, and dynamic safety threshold circle radius, it obtains the yaw angle difference of the robotic manta ray at different horizontal positions and acquires the yaw angle change. Based on the yaw angle difference, yaw angle change, and a fuzzy PID control algorithm based on the phase difference, it acquires the flapping control parameters of the robotic manta ray. Based on the flapping control parameters, it acquires the real-time phase difference corresponding to each pectoral fin of the robotic manta ray. The buoyancy acquisition module is used to assign the real-time height of the robotic manta ray as the distance component in the Z-axis direction, use the difference between the real-time height and the preset height of the fixed-point benthic task as the acquired height difference, and use the difference between the real-time height and the height at the previous moment as the height change. Based on the height change, height difference, and buoyancy, a fuzzy PID control algorithm is used to acquire the buoyancy control parameters of the robotic manta ray. The buoyancy of the robotic manta ray is acquired based on the buoyancy control parameters and the preset height of the fixed-point benthic task. And a control module for benthic control of the robotic manta ray based on its buoyancy and the real-time phase difference corresponding to each pectoral fin.

[0021] The beneficial effects of this invention are: 1. This invention decouples the three-dimensional precise benthic problem into a horizontal channel and a vertical channel, and sets an adjustable dynamic safety threshold circle with a fixed point marker as the center, so that the robotic manta ray can quickly maneuver into the dynamic safety threshold circle and complete the benthic process within the circle. Since the dynamic safety threshold circle is adjustable, it can adapt to narrow spaces of different scales and ensure the safety of precise benthic life in narrow spaces.

[0022] 2. Based on controlling the pectoral fin phase difference to adjust the heading, this invention adds a control parameter holding strategy based on exponential decay time sequence in the case of visual target loss. This enables the robotic manta ray to regain the target when the visual target is temporarily lost due to environmental disturbances, reducing the probability of permanent target loss, and also ensuring the safety of the machine itself when the target is lost.

[0023] 3. The system of the present invention adopts a modular design, with each module operating independently and communicating in real time, which has good portability and deployability, and has a small amount of computation. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating a visual servo autonomous and precise benthic control method for robotic manta rays designed for confined spaces, according to the present invention. Figure 2 This is a schematic diagram of pose information in an embodiment of the present invention; Figure 3 This is a schematic diagram of the yaw angle difference of the robotic manta ray within the dynamic safety threshold circle in an embodiment of the present invention; Figure 4 This is a schematic diagram of the yaw angle difference of the robotic manta ray outside the dynamic safety threshold circle in an embodiment of the present invention; Figure 5 This is a three-dimensional trajectory diagram of the robotic manta ray in an embodiment of the present invention; Figure 6 This is a convergence curve diagram showing the relative distance between the fixed-point benthic position and the robotic manta ray in the horizontal plane in an embodiment of the present invention. Figure 7 This is a convergence curve diagram of the relative positions of the robotic manta rays in an embodiment of the present invention; Figure 8 This is a diagram showing the convergence curve of the relative yaw angle in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] This invention provides an embodiment of a visual servo-based autonomous and precise benthic control method for robotic manta rays operating in confined spaces. In this embodiment, the three-dimensional precise benthic problem is decoupled into a horizontal channel and a height channel, and then autonomous and precise benthic control is performed. Figure 1 As shown, the control steps include: S1. Set the initialization parameters for the robotic manta ray to be stationary on the bottom; Specifically, the initialization parameters include: the reference amplitude on both sides of the pectoral fin, the reference phase difference, the flapping frequency, and the preset altitude for the fixed-point benthic mission.

[0028] For example, such as Figure 1 As shown, in one specific embodiment, initialization parameters for the robotic manta ray's precise benthic positioning need to be set in advance when the precise benthic mission in a confined space begins. In this embodiment, the initialization parameters are: the reference amplitude of the left pectoral fin. Reference amplitude of right pectoral fin Left pectoral fin baseline phase difference Right pectoral fin baseline phase difference Plosion frequency and the preset altitude for benthic missions cm.

[0029] S2. Obtain the relative pose information between the robotic manta ray and the marker at the fixed benthic location; For example, such as Figure 2 As shown, in one specific embodiment, the location marker of the fixed-point benthic position is taken as the origin of the world coordinate system, the rightward direction of the robotic manta ray is the positive X-axis, the forward direction of the robotic manta ray is the positive Y-axis, and the upward direction of the robotic manta ray is the positive Z-axis. A clockwise rotation around the Z-axis yields the positive value of the relative yaw angle. Therefore, the relative pose information includes the distance component along the X-axis between the robotic manta ray and the marker of the fixed-point benthic position. Distance component in the Y-axis direction Distance component in the Z-axis direction and relative yaw angle .

[0030] For example, such as Figure 1 As shown, in a specific embodiment, when acquiring the relative pose information between the robotic manta ray and the marker at the fixed benthic location, if the fixed benthic location is lost, a control parameter maintenance strategy is executed. The steps of executing the control parameter maintenance strategy are as follows: setting an attenuation coefficient based on the preset steady-state value of the time-varying attenuation factor at the maximum safe attenuation time threshold and the maximum safe attenuation time threshold; acquiring the time-varying attenuation factor based on the attenuation coefficient and the loss duration; acquiring the target control law based on the effective control law of the previous moment, the system preset safety command, and the time-varying attenuation factor; and using the target control law to control the robotic manta ray.

[0031] In this embodiment, the time-varying decay factor The expression is:

[0032] in, For the preset attenuation coefficient, , In order to be in The steady-state value of the time-varying decay factor at time, For the duration of loss ( ), To achieve the maximum safe hold time threshold, For the current moment, This is the moment when the visual target is lost. In this implementation case, according to the task requirements, the following is set... , , .

[0033] The expression for the target control law is:

[0034] Among them, the system has preset security instructions. That is, the effective control law for the previous moment. With system preset security commands Dynamic weighted fusion is performed. It should be noted that during this process, the weighted fusion rate increases with the duration of data loss. The increase of time-varying decay factor The driving control weights undergo time-varying transitions: In the initial stage of data loss, the system prioritizes maintaining the effective control law from the previous frame. It utilizes inertia to overcome water resistance and maintains the course correction trend; as time progresses, the weights are smoothly adjusted to the system's preset safety commands. The robot manta ray is tilted, gradually transitioning it into a hovering state. This mechanism avoids trajectory divergence caused by blindly maintaining the old commands for an extended period, and prevents orientation loss by eliminating rotational torque, thus maximizing the probability of secondary capture while ensuring safety.

[0035] S3. Obtain the real-time phase difference corresponding to each pectoral fin of the robotic manta ray; Specifically, based on the preset height of the fixed-point benthic mission, the distance component in the Z-axis direction, the height of the work area, the top radius of the work area, and the landing area radius of the work area, the radius of the dynamic safety threshold circle is obtained; the relative distance between the fixed-point benthic position and the robotic manta ray is obtained based on the distance components in the X-axis and Y-axis directions; the yaw angle difference of the robotic manta ray at different horizontal positions is obtained based on the relative distance, relative yaw angle, and the radius of the dynamic safety threshold circle, and the yaw angle change is obtained; the flapping control parameters of the robotic manta ray are obtained based on the yaw angle difference, the yaw angle change, and a fuzzy PID control algorithm based on the phase difference; and the real-time phase difference corresponding to each pectoral fin of the robotic manta ray is obtained based on the flapping control parameters.

[0036] For example, in one specific embodiment, the expression for the radius of the dynamic safety threshold circle is:

[0037] In the formula, Indicates the radius of the dynamic safety threshold circle; Indicates the top radius of the work area; Indicates the radius of the landing area of ​​the work area; Represents the distance component along the Z-axis; Indicates the preset altitude for fixed-point benthic missions; This indicates the height of the work area. In this implementation example, the height is set according to the actual work area limitations of the task. , , , In this case The radius of this dynamic safety threshold circle allows the robotic manta ray to converge its position control to a fixed point more quickly when it approaches the landing altitude.

[0038] For example, in one specific embodiment, the relative distance between the fixed-point benthic location and the robotic manta ray is:

[0039] In the formula, This indicates the relative distance between the fixed-location benthic site and the robotic manta ray; Represents the distance component along the X-axis; This represents the distance component along the Y-axis.

[0040] For example, in one specific embodiment, the steps of obtaining the difference in yaw angle of the robotic manta ray at different horizontal positions based on the relative distance, relative yaw angle, and dynamic safety threshold circle radius, and obtaining the change in yaw angle are as follows: obtaining the desired yaw angle based on the relative distance and the size of the dynamic safety threshold circle radius; using the difference between the relative yaw angle and the desired yaw angle as the yaw angle difference; and using the difference between the relative yaw angle and the yaw angle at the previous moment as the change in yaw angle.

[0041] In this embodiment, the step of obtaining the desired yaw angle based on the relative distance and the radius of the dynamic safety threshold circle is as follows: The vector angle is obtained based on the relative distance, the distance component along the X-axis, and the distance component along the Y-axis; that is, the expression for the vector angle is:

[0042] In the formula, Represents the vector angle; This indicates the relative distance between the fixed-location benthic site and the robotic manta ray; Represents the distance component along the X-axis; Represents the distance component along the Y-axis; The target angle along the route is obtained based on the vector angle. Therefore, the expression for the target angle along the route is:

[0043] In the formula, Indicates the target angle along the way.

[0044] When the relative distance is less than or equal to the radius of the dynamic safety threshold circle, the desired yaw angle is 0; when the relative distance is greater than the radius of the dynamic safety threshold circle, the target angle along the way is used as the desired yaw angle, that is, the expression for the desired yaw angle is:

[0045] In the formula, Indicates the desired yaw angle; This indicates the radius of the dynamic safety threshold circle.

[0046] For example, in one specific embodiment, the yaw angle difference yaw angle change .in, Indicates the yaw angle at the previous moment; This represents the relative yaw angle at the current moment; thus, the difference between the position and attitude control input yaw angles within the horizontal plane channel can be obtained. With yaw angle change .

[0047] For example, in one specific embodiment, the step of obtaining the flapping control parameters of the robotic manta ray based on the yaw angle difference, the yaw angle change, and the fuzzy PID control algorithm based on the phase difference is as follows: fuzzify the yaw angle difference and the yaw angle change, and input them into the fuzzy PID control algorithm based on the phase difference to output the initial flapping control parameters of the robotic manta ray. When the distance component in the Y-axis direction is less than or equal to zero, the initial flapping control parameters are adjusted. As control parameters for the flapping motion of robotic manta rays When the distance component in the Y-axis direction is greater than zero, the initial flapping control parameters are adjusted. The negative value is used as the flapping control parameter for the robotic manta ray. In this embodiment, the expression for the flapping control parameters of the robotic manta ray is:

[0048] It should be noted that this process analyzes the forward or backward maneuver required by the robotic manta ray based on the sign of the distance component in the Y-axis direction and updates the flapping control parameters (the yaw angle change direction of the forward turning maneuver under the same flapping phase difference condition is opposite to that of the backward turning maneuver).

[0049] For example, in one specific embodiment, the step of obtaining the real-time phase difference corresponding to each pectoral fin of the robotic manta ray based on the flapping control parameters is as follows:

[0050] In the formula, This indicates the real-time phase difference corresponding to the left pectoral fin; This indicates the real-time phase difference corresponding to the right pectoral fin; This indicates the PID parameter representing the phase difference on the left side; This represents the PID parameter indicating the phase difference on the right. This indicates the baseline phase difference of the left pectoral fin; This indicates the baseline phase difference of the right pectoral fin; This indicates the flapping control parameters.

[0051] Thus, the real-time phase difference of the pectoral fins is obtained. It should be noted that in order to control the robotic manta ray to swim forward or backward while turning, it is also necessary to determine the final control phase difference of the pectoral fins by combining the distance component in the Y-axis direction with the value of 0. That is, the phase difference is as follows: when the distance component in the Y-axis direction is less than or equal to zero, the robotic manta ray swims forward, and the current real-time phase difference needs to be used as the real-time phase difference of the robotic manta ray; when the distance component in the Y-axis direction is greater than zero, the robotic manta ray swims backward, and the negative value of the current real-time phase difference needs to be used as the current real-time phase difference of the robotic manta ray. Based on the current real-time phase difference, the pectoral fins on both sides can be driven to achieve position and attitude control within the horizontal channel.

[0052] S4. Obtain the buoyancy of the robotic manta ray; Specifically, the real-time altitude of the robotic manta ray is assigned as the distance component in the Z-axis direction. The difference between the real-time altitude and the preset altitude of the fixed-point benthic mission is used as the acquired altitude difference, and the difference between the real-time altitude and the altitude at the previous moment is used as the altitude change. Based on the altitude change, altitude difference, and buoyancy, a fuzzy PID control algorithm is used to obtain the buoyancy control parameters of the robotic manta ray. The buoyancy of the robotic manta ray is obtained based on the buoyancy control parameters and the preset altitude of the fixed-point benthic mission.

[0053] For example, in one specific embodiment, the real-time height of the robotic manta ray is assigned as the distance component in the Z-axis direction, i.e., the real-time height of the robotic manta ray. The height difference is then... The change in height is ,in, This represents the height of the robotic manta ray at the previous moment. The preset altitude for benthic missions.

[0054] For example, in one specific embodiment, the buoyancy control parameters of the robotic manta ray are obtained based on the height change, height difference, and buoyancy using a fuzzy PID control algorithm. The step of obtaining the buoyancy of the robotic manta ray based on the buoyancy control parameters and the preset height of the fixed-point benthic task is as follows: the height change and height difference are fuzzified and input into the buoyancy control algorithm based on buoyancy to output the buoyancy control parameters of the robotic manta ray; then the buoyancy of the robotic manta ray obtained based on the buoyancy control parameters and the preset height of the fixed-point benthic task is:

[0055] in, It's the buoyancy of the robotic manta ray. It is the maximum output buoyancy. It is the height of the work area; These are buoyancy control parameters.

[0056] S5. Implement benthic control for robotic manta rays; Specifically, benthic control of the robotic manta ray is achieved based on its buoyancy and the real-time phase difference corresponding to each pectoral fin.

[0057] In this embodiment, the real-time height is determined. The success of a precise benthic mission is determined by two conditions: whether the preset altitude of the fixed-point benthic mission, the location of the fixed-point benthic mission, and the relative distance between the robotic manta rays meet the requirements for precise benthic positioning. This determines whether the robotic manta ray continues to maneuver.

[0058] For example, in one specific embodiment, the benthic control step for the robotic manta ray based on its buoyancy and the real-time phase difference corresponding to each pectoral fin is as follows: when the real-time height is less than or equal to the preset height of the fixed-point benthic task, and the relative distance between the fixed-point benthic location and the robotic manta ray is less than or equal to the radius of the landing area of ​​the working area, the benthic control task is successful, and the phase difference of each pectoral fin of the robotic manta ray is controlled to be 0 (i.e., ), flapping frequency is 0 (i.e. =0) and the mechanical manta ray has a buoyancy of 0 (i.e. =0); when the real-time altitude is greater than the preset altitude of the fixed-point benthic mission, and the relative distance between the fixed-point benthic position and the robotic manta ray is greater than the radius of the landing area of ​​the operating area, the benthic control mission continues, and the relative yaw angle at the current moment is assigned to the relative yaw angle at the previous moment (denoted as ). The current real-time altitude is assigned to the previous real-time altitude (denoted as...). ), and store the control rate at the current moment. = ( ).

[0059] Specifically, the expression for determining whether a benthic mission is successful is:

[0060] In the formula, Indicates the radius of the landing area of ​​the work area.

[0061] A visual servo autonomous precision benthic control system for robotic manta rays designed for confined spaces includes: an initialization module, a visual information module, a phase difference acquisition module, a buoyancy acquisition module, and a control module. The initialization module sets the initialization parameters for the robotic manta ray's fixed-point benthic position, including: reference amplitude on both sides of the pectoral fins, reference phase difference, flapping frequency, and a preset height for the fixed-point benthic task. The visual information module acquires the relative pose information between the robotic manta ray and a marker at the fixed-point benthic position, including: distance components in the X-axis, Y-axis, and Z-axis directions, and the relative yaw angle. Based on the preset height, Z-axis distance component, work area height, top radius of the work area, and bottom radius of the work area, a dynamic safety threshold radius is acquired. The phase difference acquisition module acquires the position and buoyancy of the fixed-point benthic position based on the distance components in the X-axis and Y-axis directions. The relative distance between robotic manta rays is used to obtain the yaw angle difference at different horizontal positions based on the relative distance, relative yaw angle, and dynamic safety threshold radius, and to obtain the yaw angle change. The flapping control parameters of the robotic manta rays are obtained based on the yaw angle difference, yaw angle change, and a fuzzy PID control algorithm based on the phase difference. The real-time phase difference corresponding to each pectoral fin of the robotic manta ray is obtained based on the flapping control parameters. The buoyancy acquisition module assigns the real-time height of the robotic manta rays as the distance component in the Z-axis direction, uses the difference between the real-time height and the preset height of the fixed-point benthic task as the acquired height difference, and uses the difference between the real-time height and the height at the previous moment as the height change. The buoyancy control parameters of the robotic manta rays are obtained based on the height change, height difference, and a fuzzy PID control algorithm for buoyancy. The buoyancy of the robotic manta rays is obtained based on the buoyancy control parameters and the preset height of the fixed-point benthic task. The control module is used to perform benthic control on the robotic manta rays based on their buoyancy and the real-time phase difference corresponding to each pectoral fin.

[0062] The following is in conjunction with the appendix Figures 3-8 The invention is illustrated with specific simulation data: like Figure 3 As shown, the control logic of the robotic manta ray at this time is: to maneuver with the positive Y-axis as the desired heading. Figure 4 As shown, the control logic of the robotic manta ray at this time is: to calculate the desired course based on its own position and maneuver to guide itself to move within the dynamic safety threshold circle.

[0063] like Figure 5 As shown, Figure 5 The 3D trajectory diagram of the robotic manta ray shows that when the robotic manta ray moves outside the dynamic safety threshold, it can be guided back within the dynamic safety threshold by maneuvering. This verifies the convergence performance and stability of the proposed method in practical applications from a 3D trajectory perspective. Figure 6 As shown, Figure 6 This is a convergence plot showing the relative distance between the fixed-point benthic position and the robotic manta ray in the horizontal plane. The results indicate that the distance between the robotic manta ray and the z-axis gradually converges, reflecting the convergence performance and stability characteristics of the system in a two-dimensional plane. Figure 7 As shown, Figure 7 The graph shows the convergence curves of the relative positions (x, y, z) of the robotic manta ray. The convergence in the z-direction is stable, while the x and y directions show some fluctuations in the early stages of the experiment before converging rapidly. This provides a direct demonstration of the convergence and stability of the proposed method from the perspective of one-dimensional positional information. Figure 8 As shown, Figure 8 The relative yaw angle convergence curve is presented as a supplement to the trajectory and position results, verifying the system performance from the attitude control level. Experimental results show that this method can not only achieve trajectory and position convergence, but also ensure stable attitude convergence, fully demonstrating the good convergence and operational stability of this invention in actual operation.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A visual servo-based autonomous and precise benthic control method for robotic manta rays in confined spaces, characterized in that, include: Set the initialization parameters for the robotic manta ray to be stationary on the bottom. The initialization parameters include: the reference amplitude of the two sides of the pectoral fin, the reference phase difference, the flapping frequency, and the preset height for the stationary bottom task. The relative pose information between the robotic manta ray and the marker at the fixed benthic location is obtained. The relative pose information includes the distance components in the X-axis direction, Y-axis direction, and Z-axis direction, as well as the relative yaw angle between the robotic manta ray and the marker at the fixed benthic location. Based on the preset height of the fixed-point benthic task, the distance component in the Z-axis direction, the height of the work area, the top radius of the work area, and the landing area radius of the work area, the radius of the dynamic safety threshold circle is obtained. The relative distance between the fixed-point benthic position and the robotic manta ray is obtained based on the distance components in the X-axis and Y-axis directions. The yaw angle difference of the robotic manta ray at different horizontal positions is obtained based on the relative distance, relative yaw angle, and dynamic safety threshold circle radius, and the yaw angle change is also obtained. The flapping control parameters of the robotic manta ray are obtained based on the yaw angle difference, the yaw angle change, and a fuzzy PID control algorithm based on the phase difference. The real-time phase difference corresponding to each pectoral fin of the robotic manta ray is obtained based on the flapping control parameters. The real-time altitude of the robotic manta ray is assigned as the distance component in the Z-axis direction. The difference between the real-time altitude and the preset altitude of the fixed-point benthic task is used as the acquired altitude difference value, and the difference between the real-time altitude and the altitude at the previous moment is used as the altitude change value. Based on the altitude change value, altitude difference value, and buoyancy, a fuzzy PID control algorithm is used to obtain the buoyancy control parameters of the robotic manta ray. The buoyancy of the robotic manta ray is obtained based on the buoyancy control parameters and the preset altitude of the fixed-point benthic task. Benthic control of robotic manta rays is achieved based on their buoyancy and the real-time phase difference corresponding to each pectoral fin.

2. The visual servo-based autonomous and precise benthic control method for robotic manta rays in confined spaces according to claim 1, characterized in that, The expression for the radius of the dynamic safety threshold circle is: In the formula, Indicates the radius of the dynamic safety threshold circle; Indicates the top radius of the work area; Indicates the radius of the landing area of ​​the work area; Represents the distance component along the Z-axis; Indicates the preset altitude for fixed-point benthic missions; Indicates the height of the work area.

3. The visual servo-based autonomous and precise benthic control method for robotic manta rays in confined spaces according to claim 1, characterized in that, The steps for obtaining the location of the fixed-point benthic creature and the relative distance between it and the robotic manta ray based on the distance components in the X-axis and Y-axis directions are as follows: In the formula, This indicates the relative distance between the fixed-location benthic site and the robotic manta ray; Represents the distance component along the X-axis; This represents the distance component along the Y-axis.

4. The method for visual servo autonomous and precise benthic control of robotic manta rays in confined spaces according to claim 1, characterized in that, The steps for obtaining the yaw angle difference of the robotic manta ray at different horizontal positions based on relative distance, relative yaw angle, and dynamic safety threshold radius, and for obtaining the yaw angle change, are as follows: The desired yaw angle is obtained based on the relative distance and the radius of the dynamic safety threshold circle; The difference between the relative yaw angle and the desired yaw angle is taken as the yaw angle difference. The difference between the relative yaw angle and the yaw angle at the previous moment is taken as the change in yaw angle.

5. The visual servo-based autonomous and precise benthic control method for robotic manta rays in confined spaces according to claim 4, characterized in that, The steps to obtain the desired yaw angle based on the relative distance and the radius of the dynamic safety threshold circle are as follows: The vector angle is obtained based on the relative distance, the distance component in the X-axis direction, and the distance component in the Y-axis direction; Obtain the target angle along the route based on vector angles; When the relative distance is less than or equal to the radius of the dynamic safety threshold circle, the desired yaw angle is 0; when the relative distance is greater than the radius of the dynamic safety threshold circle, the target angle along the way is used as the desired yaw angle.

6. The method for visual servo autonomous and precise benthic control of robotic manta rays in confined spaces according to claim 1, characterized in that, The steps for obtaining the flapping control parameters of the robotic manta ray based on the yaw angle difference, the yaw angle change, and the fuzzy PID control algorithm based on the phase difference are as follows: The yaw angle difference and yaw angle change are fuzzified and input into the fuzzy PID control algorithm based on phase difference, and the initial flapping control parameters of the robotic manta ray are output. When the distance component in the Y-axis direction is less than or equal to zero, the initial flapping control parameters are used as the flapping control parameters for the robotic manta ray. When the distance component in the Y-axis direction is greater than zero, the negative value of the initial flapping control parameter is used as the flapping control parameter for the robotic manta ray.

7. The visual servoing autonomous and precise benthic control method for robotic manta rays in confined spaces according to claim 1, characterized in that, The steps for obtaining the real-time phase difference for each pectoral fin of the robotic manta ray based on flapping control parameters are as follows: In the formula, This indicates the real-time phase difference corresponding to the left pectoral fin; This indicates the real-time phase difference corresponding to the right pectoral fin; This indicates the PID parameter representing the phase difference on the left side; This represents the PID parameter indicating the phase difference on the right. This indicates the baseline phase difference of the left pectoral fin; This indicates the baseline phase difference of the right pectoral fin; This indicates the flapping control parameters.

8. The method for visual servo autonomous and precise benthic control of robotic manta rays in confined spaces according to claim 1, characterized in that, When acquiring the relative pose information between the robotic manta ray and the marker at the fixed benthic location, if the fixed benthic location is lost, a control parameter preservation strategy is executed. The steps for executing the control parameter preservation strategy are as follows: The attenuation coefficient is set according to the preset steady-state value of the time-varying attenuation factor at the maximum safe holding time threshold and the maximum safe holding time threshold; The time-varying decay factor is obtained based on the decay coefficient and the loss duration; The target control law is obtained based on the effective control law of the previous moment, the system preset safety instructions, and the time-varying decay factor. The target control law is then used to control the robotic manta ray.

9. The method for visual servo autonomous and precise benthic control of robotic manta rays in confined spaces according to claim 1, characterized in that, The steps for benthic control of robotic manta rays based on their buoyancy and the real-time phase difference corresponding to each pectoral fin are as follows: When the real-time altitude is less than or equal to the preset altitude of the fixed-point benthic task, and the relative distance between the fixed-point benthic position and the robotic manta ray is less than or equal to the radius of the landing area of ​​the work area, the benthic control task is successful, and the phase difference of each pectoral fin of the robotic manta ray is controlled to be 0, the flapping frequency is controlled to be 0, and the buoyancy of the robotic manta ray is controlled to be 0. When the real-time altitude is greater than the preset altitude of the fixed-point benthic mission, and the relative distance between the fixed-point benthic position and the robotic manta ray is greater than the radius of the landing area of ​​the operation area, the benthic control mission continues, assigning the current relative yaw angle to the previous relative yaw angle, assigning the current real-time altitude to the previous real-time altitude, and storing the control rate at the current moment.

10. A visual servo autonomous precision benthic control system for robotic manta rays designed for confined spaces, characterized in that, include: The initialization module is used to set the initialization parameters for the robotic manta ray to be stationary on the bottom. The initialization parameters include: the reference amplitude of the two sides of the pectoral fin, the reference phase difference, the flapping frequency, and the preset height for the stationary bottom task. The visual information module is used to acquire the relative pose information between the robotic manta ray and the marker at the fixed-point benthic location. The relative pose information includes the distance components in the X-axis direction, Y-axis direction, and Z-axis direction, as well as the relative yaw angle between the robotic manta ray and the marker at the fixed-point benthic location. Based on the preset height of the fixed-point benthic task, the distance component in the Z-axis direction, the height of the work area, the top radius of the work area, and the landing area radius of the work area, the dynamic safety threshold circle radius is acquired. The phase difference acquisition module is used to obtain the relative distance between the fixed-point benthic position and the robotic manta ray based on the distance components in the X-axis and Y-axis directions. Based on the relative distance, relative yaw angle, and dynamic safety threshold circle radius, it obtains the yaw angle difference of the robotic manta ray at different horizontal positions and acquires the yaw angle change. Based on the yaw angle difference, yaw angle change, and a fuzzy PID control algorithm based on the phase difference, it acquires the flapping control parameters of the robotic manta ray. Based on the flapping control parameters, it acquires the real-time phase difference corresponding to each pectoral fin of the robotic manta ray. The buoyancy acquisition module is used to assign the real-time height of the robotic manta ray as the distance component in the Z-axis direction, use the difference between the real-time height and the preset height of the fixed-point benthic task as the acquired height difference, and use the difference between the real-time height and the height at the previous moment as the height change. Based on the height change, height difference, and buoyancy, a fuzzy PID control algorithm is used to acquire the buoyancy control parameters of the robotic manta ray. The buoyancy of the robotic manta ray is acquired based on the buoyancy control parameters and the preset height of the fixed-point benthic task. And a control module for benthic control of the robotic manta ray based on its buoyancy and the real-time phase difference corresponding to each pectoral fin.

Images