Water-air cross-medium robot execution control method, device and equipment

Abstract

This invention relates to a method, apparatus, and device for executing control of a water-air cross-medium robot. To address potential conflicts between multiple control objectives during gliding, such as the conflict between maintaining depth and adjusting speed, a specialized control allocation algorithm is designed to decompose multiple control objectives into specific tasks and assign them to different actuators (such as tail rudders or side wings) for execution, thereby achieving effective coordinated control of depth, speed, and attitude.

Description

Technical Field

[0001] This invention belongs to the field of robot control technology, and relates to a method, device and equipment for the execution control of a water-air cross-medium robot. Background Technology

[0002] As an emerging research direction, water-air cross-medium robots have achieved some initial results, but their theoretical framework and practical applications are still under development. Current research mainly focuses on the dynamic modeling, control strategies, and structural design of cross-medium motion. For example, biomimetic cross-medium robots achieve switching between air and water by mimicking the movement characteristics of animals in nature, while rotor-type cross-medium robots utilize the vertical take-off and landing capabilities of multi-rotors to complete the transition from water surface to air. However, most of these research results focus on cross-medium switching and motion control in a single medium, and research on water surface gliding control at a constant depth remains insufficient.

[0003] During water gliding, the fluid environment is complex, involving the adjustment of multiple controlled parameters and the output of multiple actuators, making it difficult to ensure system stability. Current control methods are mainly designed for stability underwater or in the air, lacking effective technical solutions for the control allocation and dynamic adjustment required for water surface gliding at a constant depth, and failing to achieve efficient allocation under multi-objective control. Summary of the Invention

[0004] To address the problems existing in the above-mentioned traditional technologies, this invention proposes a water-air cross-medium robot execution control method, a water-air cross-medium robot execution control device, and a computer device, which can achieve efficient allocation under multi-objective control.

[0005] To achieve the above objectives, the embodiments of the present invention adopt the following technical solutions:

[0006] On the one hand, a method for controlling a water-air cross-medium robot is provided, including the following steps:

[0007] A dynamic model based on component decomposition is established for the water-air transmedia robot in the gliding state. The forces acting on the water-air transmedia robot in the gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface when gliding underwater, the lift generated by the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy.

[0008] The dynamic model is transformed into a nonlinear least squares optimization problem. The LM algorithm is used to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot, and the thrust and servo deflection angle of each actuator of the water-air cross-medium robot are obtained.

[0009] Each input command is generated based on the thrust of each actuator and the deflection angle of the servo motor, and output to the corresponding actuator. The actuators then control the water-air transmedia robot.

[0010] On the other hand, a water-air cross-medium robot execution control device is also provided, comprising:

[0011] The model building module is used to build a dynamic model of the water-air cross-medium robot in the gliding state based on component decomposition. The forces acting on the water-air cross-medium robot in the gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface when gliding underwater, the lift generated by the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy.

[0012] The problem-solving module is used to convert the dynamic model into a nonlinear least squares optimization problem. The LM algorithm is used to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot, and to obtain the thrust and servo deflection angle of each actuator of the water-air cross-medium robot.

[0013] The command control module is used to generate input commands based on the thrust of each actuator and the deflection angle of the servo motor, and output them to the corresponding actuators to control the water-air transmedia robot.

[0014] In another aspect, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described water-air transmedia robot execution control method.

[0015] One of the above technical solutions has the following advantages and beneficial effects:

[0016] The aforementioned water-air cross-medium robot execution control method, device, and equipment address potential conflicts between multiple control objectives during gliding, such as the conflict between maintaining depth and adjusting speed. They employ a specialized control allocation algorithm to decompose multiple control objectives into specific tasks and assign them to different actuators (such as tail rudders or side wings) for execution, thereby achieving effective coordinated control of depth, speed, and attitude. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the conventional technology, the drawings used in the description of the embodiments or the conventional technology 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.

[0018] Figure 1This is a flowchart illustrating the control method for a water-air cross-medium robot in one embodiment;

[0019] Figure 2 This is a schematic diagram of the body coordinate system in one embodiment;

[0020] Figure 3 A block diagram of the controller design in one embodiment (TD part);

[0021] Figure 4 A block diagram of the controller design (PID part) in one embodiment;

[0022] Figure 5 This is a schematic diagram of the overall force analysis of the system in one embodiment;

[0023] Figure 6 This is a schematic diagram of the force analysis of a single component in one embodiment;

[0024] Figure 7 This is a schematic diagram of the module framework of the water-air cross-medium robot execution control device in one embodiment. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0026] It should be noted that, in this document, the reference to "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The presentation of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will understand that the embodiments described herein can be combined with other embodiments. The term "and / or" as used herein refers to any combination of one or more of the associated listed items, and all possible combinations, including such combinations.

[0027] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0028] A water-air cross-medium robot (UAUV) is an unmanned aerial vehicle capable of adaptively transitioning between two different fluid media, underwater and air, and autonomously navigating continuously. It breaks the limitations of previous unmanned aerial vehicles (UAVs), unmanned surface vessels (USVs), and unmanned underwater vehicles (UUVs) that could only navigate in a single specific environment, achieving the goal of simultaneously conducting aerial, surface, and underwater exploration of a specific area using a single unit. Existing water-air cross-medium robots include biomimetic, rotorcraft, fixed-wing, and hybrid UAUV prototypes, which achieve switching between air and water by mimicking the movement characteristics of animals in nature or utilizing the vertical takeoff and landing capabilities of multi-rotors. However, most of these existing research results focus on cross-medium switching and motion control in a single medium, lacking in-depth research on surface gliding control at a constant depth.

[0029] Existing small current meters for underwater robots have insufficient measurement range, failing to meet the practical needs of measuring the gliding speed of water-to-air cross-medium robots. Considering the limitations of current small current meters in terms of measurement speed range—for example, the maximum measurable speed of the DVL-A50 Doppler velocimeter is only 3.75 m / s, while the DVL-A125 Doppler velocimeter, although increasing the maximum measurable speed to 9 m / s, is relatively large and heavy, unsuitable for the hydrofoil gliding scenario of water-to-air cross-medium robots—this specification proposes a gliding speed estimation algorithm based on a tracking differentiator (TD) to accurately estimate the speed of water-to-air cross-medium robots during hydrofoil gliding.

[0030] Existing small current meters used for underwater robots have insufficient measurement range and cannot meet the actual needs of gliding speed measurement. Moreover, they lack control methods for gliding at a constant depth on the water surface: existing control algorithms are mainly designed for cross-medium switching or single-medium motion, which is difficult to meet the motion requirements of multiple disturbances and multiple controlled parameters during gliding on the water surface.

[0031] To address the aforementioned problems, this invention proposes innovative solutions from two aspects: Firstly, it proposes a water surface gliding speed estimation algorithm based on a tracking differentiator (TD) to accurately estimate the speed of a water-air cross-medium robot during water surface gliding; secondly, for the control requirements of hydrofoils gliding at constant depth and speed, it designs a water surface gliding cooperative control algorithm based on an extended state observer (ESO) that integrates PID and multiple control output allocation. The extended state observer estimates and compensates for environmental disturbances and internal system disturbances (including nonlinear coupling components), achieving accurate control at constant depth and speed. This water surface gliding cooperative control algorithm provides an innovative solution for the motion control of water-air cross-medium robots.

[0032] Among them, the dual-loop PID (Proportional-Integral-Derivative) controller is a commonly used feedback control method. It adjusts the system deviation in real time through three stages: proportional (P), integral (I), and derivative (D). Proportional control is used for rapid response to deviation, integral control is used to eliminate steady-state errors, and derivative control is used to predict the trend of deviation changes and prevent overshoot. The dual-loop PID controller is based on a multi-stage PID control structure. The outer loop is mainly used for target control of position or velocity, while the inner loop is responsible for rapid response control of attitude angle or angular velocity.

[0033] Extended State Observer (ESO): An advanced state estimation technique used for real-time estimation of system state variables and external disturbances. Compared to traditional state observers, ESO can not only estimate internal system state variables but also estimate and compensate for unknown external disturbances. Its core design concept is to treat the unknown disturbance of the system as an "extended state" and incorporate it into the observer design. By designing appropriate observer gain, ESO can quickly and accurately estimate the system state and disturbances and feed this information back to the controller, thereby improving the robustness and anti-interference capability of the control system. Levenberg-Marquardt (LM) Algorithm: An optimization algorithm for solving nonlinear least squares problems, an improvement on the Gauss-Newton method. The LM algorithm combines the advantages of the steepest descent method and the linearization method (Taylor series expansion), and is applicable to different stages where parameter estimates are far from and close to the optimal value, thus finding the optimal solution faster.

[0034] The water-air cross-medium robot (hereinafter referred to as the system) is subjected to a combination of forces and torques during its gliding on the water surface. These forces include the thrust provided by the thrusters, the water resistance encountered by the control surfaces during gliding underwater, the lift generated by the difference in fluid velocity above and below the control surfaces, the drag generated by other components, gravity, and buoyancy. The resultant force and resultant torque of these forces affect the system's gliding depth, gliding speed, and attitude in the water. The controlled parameters of the system include gliding speed, gliding depth, and attitude angles (such as roll angle and pitch angle). The gliding speed is estimated by a water surface gliding speed estimation algorithm based on a tracking differentiator and is determined by the resultant force along the system's forward direction; the gliding depth is directly measured by the system's existing water level gauge and is determined by the resultant force along the vertical direction of the system's plane; the attitude information is obtained from the inertial measurement unit (IMU) in the system's flight control system, and its pitch angle is measured. With roll angle The torques are determined by the system in the pitch and roll directions, respectively. The core design of the controller is to use the deviations in taxiing speed, taxiing depth, and attitude angle as inputs, and then solve for the forces and torques required to be applied to each axis of the hydrofoil as outputs.

[0035] Since the results calculated by each controller are the desired forces and torques of each axis of the hydrofoil, in order to map these desired values ​​to the specific inputs of the hydrofoil actuators (i.e., the deflection angle of the control surfaces and the thrust of the underwater propulsion), a control allocation algorithm needs to be designed to achieve this goal. This control allocation algorithm can reasonably decompose and allocate the desired forces and torques to each actuator to ensure the stability and control effect of the hydrofoil's gliding.

[0036] In one embodiment, such as Figure 1 As shown, a method for controlling a water-air cross-medium robot is provided, which may include the following steps S12 to S16:

[0037] S12, Establish a dynamic model of the water-air cross-medium robot in gliding state based on component decomposition; the forces on the water-air cross-medium robot in gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface when gliding underwater, the lift generated by the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity and buoyancy.

[0038] S14. The dynamic model is transformed into a nonlinear least squares optimization problem. The LM algorithm is used to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot, and the thrust and servo deflection angle of each actuator of the water-air cross-medium robot are obtained.

[0039] S16 generates input commands based on the thrust of each actuator and the deflection angle of the servo motor, and outputs them to the corresponding actuators to control the water-air transmedia robot.

[0040] Understandably, for clarity in the subsequent description, a body coordinate system has been defined, such as... Figure 2 As shown. This coordinate system has its origin at the center of gravity of the water-air transmedia robot system. . The axis runs along the direction of the fuselage's movement, from the tail to the nose; The shaft lies within the plane of the machine body and is perpendicular to it. The axis points to the right wing; The axis runs perpendicular to the plane of the fuselage and... shaft and The axes form a right-handed coordinate system.

[0041] The gliding speed estimation process is as follows: a set of differential equations is used to track the reference position and dynamically estimate the gliding speed. The gliding speed estimation algorithm uses the reference position of the water-air transmedia robot during the hydrofoil gliding process. The input is used as the input, and adjustments are made based on error feedback regarding the hydrofoil's position during gliding. The final output consists of two velocity components representing the gliding velocity of the water-air transmedia robot. These two components correspond to respectively as The estimated velocity along the axial direction and Estimated velocity along the axial direction.

[0042] Furthermore, the taxiing speed estimation algorithm uses the following differential equation (state-space model of a second-order linear time-invariant system) to estimate the taxiing speed:

[0043] ;

[0044] In the body coordinate system, This is the reference position of the hydrofoil on the water surface. They are estimated respectively Axial direction and Position along the axis. and The estimated water-air transmedia robot is in The estimated velocity along the axial direction and Estimated velocity along the axial direction. These are control parameters (which can be set according to the position tracking error requirements in specific application scenarios) used to adjust the system's tracking speed for position errors. for The rate of change of position in the axial direction for Rate of change of axial velocity for The rate of change of position in the axial direction for The rate of change of velocity in the axial direction. Similarly, the top dot of the parameter in the following text represents the first derivative.

[0045] The gliding speed estimation algorithm is based on the input water surface reference position. By dynamically adjusting the system state through a set of differential equations, the two components of the gliding velocity of the water-air transmedia robot are ultimately output. Compared to traditional flowmeter measurement methods, the above-mentioned method for measuring the surface gliding speed of a water-air transmedia robot can accurately estimate the hydrofoil gliding speed of the robot without additional hardware burden, and has high real-time performance and adaptability.

[0046] In one embodiment, the controller of the water-air transmedia robot includes a gliding speed controller, a gliding depth controller, a pitch angle controller, and a roll angle controller.

[0047] like Figure 3 The red box in the controller of the water-air cross-medium robot shows the location of the tracking differentiator TD, indicating its position information. First, the data is processed by a tracking differentiator TD, which outputs an estimate of the hydrofoil's current gliding speed. Next, the estimated current gliding speed of the hydrofoil is... With the expected speed of the hydrofoil The deviation between the input and output values ​​is fed into the PID controller (which, together with the tracking differentiator TD, forms the coasting speed controller). The PID controller can then calculate and output the coasting speed based on this deviation. Desired force in the axial direction . For linearization processing along Desired force in the axial direction, For the desired depth, To represent the actual measured depth, The desired angle for pitch. The actual pitch angle measured in real time by the inertial measurement unit (IMU). This is the estimated value of the disturbance in the depth direction. The pitch angle deviation is processed by a single-stage PID controller to obtain the desired Euler angular velocity. This is the Euler angular velocity transformation matrix. Let ω be the desired angular velocity in the body coordinate system, and FF be the feedforward control unit. For the expected roll angle, Indicates roll angle The time derivative, Indicates roll angle The measured value, The measured value representing the roll rate. This is the expected value of the roll rate. Roll angle The expected value of the perturbation, The measured value representing the pitch rate. Let the pitch angle be the expected value of the disturbance. m d This represents the output value of the pitch controller. This is the measured value of the torque in the pitch direction. This indicates the output value of the roll angle controller. This indicates the measured torque value in the roll direction. This indicates the desired thrust of the left thruster. This represents the desired thrust of the right-side thruster. This indicates the desired angle of the left hydrofoil. This indicates the desired angle of the right hydrofoil. Indicates the desired horizontal tail angle.

[0048] ESO fusion PID (such as Figure 4 (Module A in the text): The core design of the controller lies in using the deviation of the controlled parameters (such as taxiing speed, taxiing depth, and attitude angle) as input, and mapping it to the forces and torques required to be applied to each axis of the hydrofoil under appropriate parameters. This embodiment uses a PID controller with feedforward control and innovatively integrates an extended state observer (ESO) to compensate for disturbances, thereby obtaining a more stable and accurate control effect.

[0049] The system's gliding depth in water is affected by both the direction of travel and the speed of travel. The force in the axial direction is also affected by the moment in the pitch direction. Therefore, as Figure 4 As shown, the deviation in gliding depth (Based on expected depth) Compared with the actual measured depth (The result obtained by subtraction) is, on the one hand, linearized and used as the path. Desired force in the axial direction On the other hand, after processing by a first-stage PID controller, the output of the PID controller is limited and used as the desired pitch angle. .

[0050] Specifically, the dynamic characteristics of the height portion can be described by the following equation:

[0051] (1)

[0052] in, This represents the perturbation term in the depth direction. (Take...) ,have:

[0053] (2)

[0054] The control law for designing the gliding depth is:

[0055] (3)

[0056] Substituting equation (3) of the control law for the gliding depth into equation (1), we obtain the following gliding depth controller:

[0057] (4)

[0058] Right now:

[0059] (5)

[0060] in, For control parameters, t For time, For the estimated depth-direction perturbation, For the system speed.

[0061] The pitch control B employs a cascaded design of outer and inner loops, such as... Figure 4 As shown. The outer ring uses the pitch angle as the controlled parameter, and the desired angle of the pitch angle is determined by... The actual pitch angle measured in real time by the IMU The pitch angle deviation is calculated by subtraction. This pitch angle deviation is then processed by a single-stage PID controller to obtain the desired Euler angular velocity. .

[0062] The kinematics based on Euler angles can be expressed as:

[0063] (6)

[0064] Pick ,have:

[0065] (7)

[0066] The control law for the pitch angular velocity is designed as follows:

[0067] (8)

[0068] According to the control law (8) of pitch angular velocity, the pitch angular velocity error term can be obtained:

[0069] (9)

[0070] in, The pitch angular velocity, These are control parameters.

[0071] Next, using the Euler angular velocity transformation matrix The desired Euler angular velocity Desired angular velocity converted to body coordinates This expected angular velocity The input is fed into the inner-loop PID controller, and its output is added to the output of the feedforward (FF) loop to form the desired torque of the system in the pitch direction. The data is then passed to the Control Allocator for further allocation processing.

[0072] Based on the relationship between Euler's angular velocity and the body's angular velocity, we can conclude that:

[0073] (10)

[0074] The pitch dynamics formula is:

[0075] (11)

[0076] Pick ,have:

[0077] (12)

[0078] The control law for pitch angular velocity in the body coordinate system is designed as follows:

[0079] (13)

[0080] Substituting the control law (13) into the pitch dynamics formula (12), we obtain the pitch angle controller in the body coordinate system:

[0081] (14)

[0082] in, Yaw angle They are respectively around , , Moment of inertia in the axial direction, For the inertia product, r For the coordinate system around the body angular velocity of the axis, q For the coordinate system around the body angular velocity of the axis, m This is the actual torque in the pitch direction. For the disturbance term, For the estimated disturbance term, For control parameters. The constants appearing in formula (11) are calculated by the following formula:

[0083] (15)

[0084] The design of the roll control C is the same as that of the pitch control, but the desired roll angle is different. It is given via remote control. The output of the roll angle controller then serves as the desired torque of the system in the roll direction. .

[0085] In one embodiment, the taxi depth controller includes an extended state observer for disturbance compensation of the depth loop, and both the pitch and roll angle controllers include extended state observers for disturbance compensation of the angular rate loop.

[0086] It is understandable that the system is subject to disturbances from water flow and airflow during gliding on the water surface, making it difficult to achieve ideal control results using only a traditional PID controller. Therefore, this embodiment innovatively integrates an Extended State Observer (ESO) into the depth and angular rate loop PID controllers, effectively compensating for external disturbances and improving the system's robustness and control accuracy. The ESO takes the control input and output of the controlled object as inputs, and its output includes three parts: an estimated value of the controlled object's output, the first derivative of that estimate, and the second derivative. The second derivative of the estimated value of the controlled object's output is fed back to the PID controller output for real-time compensation of system disturbances, thereby enhancing control performance.

[0087] Specifically, in the depth control part, the equation for ESO can be written according to formula (1) as follows:

[0088] (16)

[0089] In the angular rate control section, the equation for ESO can be written according to formula (12) as follows:

[0090] (17)

[0091] Through the above ESO design, the system can estimate and compensate for external disturbances in real time. For example, in deep control, As a disturbance estimate, it is fed back to the PID controller output to compensate for disturbances; while in angular rate control, As a disturbance estimate, it is also fed back to the PID controller output for real-time compensation of system disturbances.

[0092] Thus, the proposed water gliding speed estimation algorithm and water gliding cooperative control algorithm significantly improve the stability and control performance of UAUVs during water gliding. These technological advancements not only address the limitations of sensor performance in existing technologies but also provide an effective gliding speed measurement scheme for small water-air cross-medium robots. The aforementioned algorithms possess high real-time performance and adaptability, enabling them to operate accurately in complex fluid environments, which is crucial for improving the robot's operational efficiency.

[0093] By adjusting the pitch angle, the water-air cross-medium robot can flexibly control its gliding depth. Adjusting the pitch angle is equivalent to changing the angle between the hydrofoil or control surface and the water flow, thereby changing the magnitude of the force and keeping the water-air cross-medium robot within a set depth range. Secondly, the speed and attitude during gliding are also coordinated and controlled. The above technical solution fully utilizes the advantages of algorithms under existing hardware conditions, solving the bottleneck problem of speed measurement and improving the stability of the water-air cross-medium robot during gliding. Although current sensor technology has not yet fully resolved the contradiction between miniaturization and high precision, with future hardware technology development, such as the development of lighter and higher-precision speed sensors, the overall performance and application prospects of the above technical solution will be further expanded.

[0094] Furthermore, model-based control allocation:

[0095] During the gliding process of a trans-medium water robot, multiple controlled parameters need to be adjusted to achieve stable control, such as gliding speed, gliding depth, and attitude angles (including pitch and roll angles). The realization of these controlled parameters depends on the coordinated operation of various actuators on the trans-medium water robot. However, the forces and torques between these actuators are coupled; for example, the deflection of the horizontal stabilizer simultaneously affects both the pitch angle and gliding depth; the lift generated by the hydrofoils not only determines the stability of the gliding depth but may also cause changes in the roll angle. This coupling characteristic of the forces between actuators makes control allocation a critical aspect of system control.

[0096] In water gliding control, each controller (such as the gliding speed controller, gliding depth controller, pitch angle controller, and roll angle controller) calculates the target force or torque required by the water-air transmedium robot. These target quantities need to be ultimately achieved through actuators such as control surfaces, hydrofoils, and thrusters. However, due to the interaction between the actuators, it is impractical to simply map the controller output directly to a single actuator. For example, adjusting the horizontal stabilizer alone may not fully meet the pitch angle control requirements, or the thrust output of the thruster may simultaneously affect speed and attitude stability. Therefore, a control allocation algorithm is needed to rationally decompose the target force and torque according to the control requirements and the dynamic characteristics of the actuators to generate specific input commands for each actuator.

[0097] Control assignment algorithms not only need to address the force coupling problem between actuators, but also need to handle the physical constraints of the actuators and the nonlinear characteristics of the system. In practical applications, the deflection angle of the control surface, the lift of the hydrofoil, and the thrust of the propeller all have their own operating ranges. Control assignment needs to ensure that the generated commands are within the operating range of the actuators. Furthermore, due to the nonlinear dynamics of robot gliding control, control assignment must also consider the complex nonlinear relationship between the actuators and the control target to ensure the efficiency and accuracy of command assignment.

[0098] To achieve the aforementioned control allocation objectives, this embodiment proposes a component-decomposition-based dynamic model for coasting. Based on this dynamic model, a control allocation algorithm for coasting is further designed to achieve efficient allocation under multi-objective control.

[0099] The symbols used in the dynamic model are explained in Table 1.

[0100] Table 1

[0101]

[0102] like Figure 5 As shown, let In coordinate system The coordinates are thrust of the propeller The point of application in the coordinate system The coordinates below are ,thrust The point of application in the coordinate system The coordinates below are Aerodynamics The point of application (center of pressure) in the coordinate system The coordinates below are aerodynamics The point of application (center of pressure) in the coordinate system The coordinates below are Then the thrust The point of application in the coordinate system The coordinates below are ,thrust The point of application in the coordinate system The coordinates below are Aerodynamics The point of application in the coordinate system The coordinates below are Aerodynamics The point of application in the coordinate system The coordinates below are .

[0103] To further describe the dynamic characteristics of the system during taxiing, it is necessary to analyze the combined effects of various forces and moments acting on the system during taxiing. These forces include the thrust provided by the propeller, the water resistance experienced by the control surface during underwater taxiing, the lift generated by the difference in fluid velocity above and below the control surface, the drag generated by other components, gravity, and buoyancy. These forces collectively determine the system's motion state and form the basis of dynamic modeling. Based on these forces, the system's dynamic equations are established from both force and moment perspectives, as follows:

[0104] (18)

[0105] (19)

[0106] in, , , This indicates the lift generated by the control surfaces. This indicates the drag generated by the control surface. , .like Figure 6 As shown, assuming that the lift coefficient and drag coefficient of the wing surface are only related to the angle of attack, then:

[0107] (20)

[0108] Rearranging equation (18), we get:

[0109] (twenty one)

[0110] Reconstruct the complex matrix into a product of the coefficient matrix and a vector containing only force-torque variables:

[0111] (twenty two)

[0112] In practical applications, to simplify calculations and analysis, it is usually assumed that the point of action of the motor and the point of action of the hydrodynamic force coincide, i.e. , Under this assumption, the dynamic equations can be further simplified to the following form:

[0113] (twenty three)

[0114] By combining terms with the same coefficient, the matrix representation can be further simplified:

[0115] (twenty four)

[0116] Combining equations (17) and (20), we get:

[0117] (25)

[0118] Considering that the horizontal stabilizer mainly generates pitching moment during gliding on the water surface Meanwhile, the drag generated by the fuselage and other components is ignored. The effects are as follows:

[0119] (26)

[0120] The matrices and vectors in equation (25) are symbolically represented, where the force and torque vectors on the left are... Recorded as , representing the total force and torque vector of the system; the coefficient matrix is ​​denoted as . , representing the mapping relationship between the total force and torque acting on the system and the force and torque output by the actuator; the vector on the right side of the coefficient matrix is ​​denoted as , representing the resultant force and resultant torque output by the actuator; vector Recorded as This represents the additional external forces and moments of the system, including gravity. ,buoyancy and the counter-torque generated by the motor .

[0121] Based on the above symbol definitions, equation (25) can be simplified as:

[0122] (27)

[0123] The controller processes the additional external forces and torques of the system; that is, the desired force and torque output by the controller has already been subtracted from the additional external forces and torques. Therefore:

[0124] (28)

[0125] Combining equations (26) and (27), we get:

[0126] (29)

[0127] Therefore, the expected output resultant force and resultant moment vector of the actuator are:

[0128] (30)

[0129] The desired output resultant force and resultant torque of the actuator can be calculated using equation (29). According to equation (25), the relationship between the output resultant force of the actuator, the actuator thrust, and the control surface deflection angle is as follows:

[0130] (31)

[0131] Finally, the thrust of each actuator needs to be calculated. and servo deflection angle However, since the expression in equation (30) contains nonlinearity involving sine and cosine, it is difficult to solve directly. Therefore, the problem in equation (30) is transformed into the following nonlinear least squares optimization problem for solution:

[0132] (32)

[0133] in,

[0134] (33)

[0135] The thrust and servo deflection angle of each actuator can be obtained by solving this nonlinear least squares problem using the LM algorithm, thereby achieving the final control allocation.

[0136] The aforementioned water-air cross-medium robot execution control method addresses the potential conflicts between multiple control objectives during gliding, such as the conflict between maintaining depth and adjusting speed. It designs a specialized control allocation algorithm to decompose multiple control objectives into specific tasks and assign them to different actuators (such as tail rudders or side wings) for execution, thereby achieving effective coordinated control of depth, speed, and attitude.

[0137] In some implementations, the problem of estimating gliding speed can be solved through both algorithmic and hardware alternatives. Optionally, the water surface gliding speed estimation algorithm proposed in this invention mainly addresses the problem that current sensors cannot simultaneously achieve small size, light weight, and large measurement range. For algorithmic alternatives, classic signal processing methods such as Kalman filtering or particle filtering can also be used. These methods can achieve accurate gliding speed estimation even when sensor data contains noise or uncertainty. The specific speed estimation process can be understood by referring to the existing calculation processes of these two classic signal processing methods, and will not be elaborated further in this embodiment. Meanwhile, regarding hardware solutions, although there is currently no flow velocity measuring instrument that can simultaneously achieve small size, light weight, and large measurement range, this challenge can be overcome through future advancements in hardware technology. For example, higher precision and lighter flow velocity sensors can be developed, such as improved laser Doppler current meters or ultrasonic velocimeters. These devices may further optimize size and weight while maintaining a large measurement range.

[0138] In some implementations, alternatively, for the waterplane gliding cooperative control algorithm, the depth control portion can also use existing fuzzy control algorithms, model predictive control (MPC), or sliding mode control as alternatives. These methods can also achieve the control objective of multivariable coupling characteristics in hydrofoil gliding. The control allocation algorithm can also be replaced by optimal control methods (such as existing linear quadratic regulators (LQRs) or constraint optimization-based allocation methods), which can also be used to achieve cooperative control among multiple controlled parameters.

[0139] Compared to traditional technologies, the aforementioned technical solutions demonstrate broad application prospects in various types of navigation applications. For example, UAUVs can perform underwater reconnaissance and surveillance missions, while possessing rapid deployment and air mobility capabilities. In marine scientific research, UAUVs can conduct marine data collection, environmental monitoring, and seabed exploration, providing more efficient and economical solutions. In maritime search and rescue operations, UAUVs can quickly respond to maritime emergencies, conduct large-scale searches and rescues, and improve search and rescue efficiency and success rates. In the field of resource development, UAUVs provide precise data support and operational assistance in offshore oil and gas exploration and seabed mineral development. In environmental monitoring, UAUVs can be used to monitor the impact of marine pollution and climate change on marine ecosystems, providing scientific basis for environmental protection.

[0140] It should be understood that, although the above process Figure 1 The steps in the diagram are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order in which these steps are executed; they can be performed in other orders. Furthermore, the above process... Figure 1 At least some of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0141] In one embodiment, such as Figure 7As shown, a water-air cross-medium robot execution control device 100 is also provided, including a model building module 11, a problem-solving module 13, and a command control module 15. The model building module 11 is used to build a dynamic model of the water-air cross-medium robot in a gliding state based on component decomposition. The forces acting on the water-air cross-medium robot in the gliding state in the dynamic model include the thrust provided by the thrusters, the water resistance experienced by the rudder surface during underwater gliding, the lift generated by the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy. The problem-solving module 13 is used to convert the dynamic model into a nonlinear least squares optimization problem, and uses the LM algorithm to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot to obtain the thrust of each actuator and the rudder deflection angle. The command control module 15 is used to generate input commands based on the thrust and rudder deflection angle of each actuator and output them to the corresponding actuators, thereby controlling the water-air cross-medium robot through the actuators.

[0142] The aforementioned water-air cross-medium robot execution control device 100 addresses potential conflicts between multiple control objectives during gliding, such as the conflict between maintaining depth and adjusting speed. It employs a specialized control allocation algorithm to decompose multiple control objectives into specific tasks and assign them to different actuators (such as tail rudders or side wings) for execution, thereby achieving effective coordinated control of depth, speed, and attitude.

[0143] In one embodiment, the controller of the water-air transmedia robot includes a gliding speed controller, a gliding depth controller, a pitch angle controller, and a roll angle controller.

[0144] In one embodiment, the taxi depth controller includes an extended state observer for disturbance compensation of the depth loop, and both the pitch and roll angle controllers include extended state observers for disturbance compensation of the angular rate loop.

[0145] It is understood that the explanations of the features in the aforementioned water-air cross-medium robot execution control device 100 can be understood by referring to the corresponding explanations in the various embodiments of the aforementioned water-air cross-medium robot execution control method. Each module in the aforementioned water-air cross-medium robot execution control device 100 can be implemented entirely or partially through software, hardware, or a combination thereof. The aforementioned components can be embedded in hardware or independently of a device with data processing capabilities, or stored in software in the memory of the aforementioned device, so that the processor can call and execute the operations corresponding to each module. The aforementioned device can be, but is not limited to, various types of computers already existing in the art.

[0146] In one embodiment, a computer device is also provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following processing steps: establishing a component-decomposition-based dynamic model of the water-air cross-medium robot in a gliding state; the forces acting on the water-air cross-medium robot in the gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface during gliding underwater, the lift generated due to the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy; converting the dynamic model into a nonlinear least squares optimization problem, and using the LM algorithm to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot to obtain the thrust and rudder deflection angle of each actuator of the water-air cross-medium robot; generating input commands based on the thrust and rudder deflection angle of each actuator and outputting them to the corresponding actuators, thereby controlling the water-air cross-medium robot through each actuator.

[0147] In one embodiment, when the processor executes the computer program, it can also implement the steps or sub-steps added in the various embodiments of the above-described water-air transmedia robot execution control method.

[0148] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the following processing steps are implemented: establishing a dynamic model of the water-air cross-medium robot in a gliding state based on component decomposition; the forces acting on the water-air cross-medium robot in the gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface when gliding underwater, the lift generated due to the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy; converting the dynamic model into a nonlinear least squares optimization problem, and using the LM algorithm to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot to obtain the thrust and rudder deflection angle of each actuator of the water-air cross-medium robot; generating input commands based on the thrust and rudder deflection angle of each actuator and outputting them to the corresponding actuators, thereby controlling the water-air cross-medium robot through each actuator.

[0149] In one embodiment, when the computer program is executed by the processor, it can also implement the steps or sub-steps added in the various embodiments of the above-described water-air cross-medium robot execution control method.

[0150] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), memory bus DRAM (RDRAM), and interface DRAM (DRDRAM), etc.

[0151] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0152] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of protection of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and all such modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for executing control of a water-air cross-medium robot, characterized in that, Including the following steps: A dynamic model based on component decomposition is established for the water-air transmedia robot in the gliding state. The forces acting on the water-air transmedia robot in the gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface when gliding underwater, the lift generated by the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy. The dynamic model is transformed into a nonlinear least squares optimization problem. The LM algorithm is used to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot, and the thrust and servo deflection angle of each actuator of the water-air cross-medium robot are obtained. Each input command is generated based on the thrust of each actuator and the deflection angle of the servo motor, and output to the corresponding actuator. The actuators then control the water-air transmedia robot.

2. The water-air cross-medium robot execution control method according to claim 1, characterized in that, The controller for the water-air cross-medium robot includes a gliding speed controller, a gliding depth controller, a pitch angle controller, and a roll angle controller.

3. The water-air cross-medium robot execution control method according to claim 2, characterized in that, The taxi depth controller includes an extended state observer for disturbance compensation in the depth loop, and both the pitch and roll controllers include extended state observers for disturbance compensation in the angular rate loop.

4. A water-air cross-medium robot execution control device, characterized in that, include: The model building module is used to build a dynamic model of the water-air cross-medium robot in the gliding state based on component decomposition. The forces acting on the water-air cross-medium robot in the gliding state in the dynamic model include the thrust provided by the thruster, the water resistance experienced by the rudder surface when gliding underwater, the lift generated by the difference in fluid velocity above and below the rudder surface, the resistance generated by other components, gravity, and buoyancy. The problem-solving module is used to convert the dynamic model into a nonlinear least squares optimization problem. The LM algorithm is used to solve the nonlinear least squares optimization problem based on the controller of the water-air cross-medium robot, and to obtain the thrust and servo deflection angle of each actuator of the water-air cross-medium robot. The command control module is used to generate input commands based on the thrust of each actuator and the deflection angle of the servo motor, and output them to the corresponding actuators to control the water-air transmedia robot.

5. The water-air cross-medium robot execution control device according to claim 4, characterized in that, The controller for the water-air cross-medium robot includes a gliding speed controller, a gliding depth controller, a pitch angle controller, and a roll angle controller.

6. The water-air cross-medium robot execution control device according to claim 5, characterized in that, The taxi depth controller includes an extended state observer for disturbance compensation in the depth loop, and both the pitch and roll controllers include extended state observers for disturbance compensation in the angular rate loop.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the water-air transmedia robot execution control method as described in any one of claims 1 to 3.

Images