A tandem dual-rotor unmanned aerial vehicle and its motion control method

Abstract

This application provides a tandem dual-rotor unmanned aerial vehicle (UAV) and its motion control method. By acquiring the UAV's flight altitude and comparing it with preset first and second altitude thresholds, the method accurately identifies the UAV's flight state and calculates the medium immersion ratio coefficient in real time. Simultaneously, a timed constraint mechanism is used to achieve smooth switching between the thrusters and the controller. Real-time UAV status data is then collected, and a LOS guidance law model and an NTSTSM control law model are constructed by combining the desired trajectory data. Smooth calculations are performed with the immersion ratio coefficient as the core, and continuous and stable thruster control data is output. This ensures that the force and control commands of the UAV do not change abruptly during the water-air cross-medium process, achieving a smooth transition of buoyancy, fluid damping, and added mass. This effectively avoids the risk of fuselage shaking and capsizing caused by sudden force changes during cross-medium travel. At the same time, the timed switching constraint mechanism eliminates frequent switching between the air and underwater controllers caused by sea surface fluctuations, ensuring control continuity and attitude stability.

Description

Technical Field

[0001] This application relates to the field of cross-media unmanned aerial vehicle (UAV) technology, and in particular to a tandem dual-rotor UAV and its motion control method. Background Technology

[0002] A cross-medium unmanned vehicle is a vehicle that can fly in the air and dive underwater. It has the ability to move quickly and over a wide area and can be used to perform tasks such as marine resource exploration, water topography surveying and maritime rescue. It also has a wide range of applications in the defense field.

[0003] When current cross-medium unmanned aerial vehicles (UAVs) cross the water-air interface, the physical properties of the medium change abruptly, causing buoyancy, fluid damping, and added mass to change drastically in an instant. This can easily lead to attitude instability, fuselage shaking, or even capsizing. At the same time, repeated high-altitude oscillations near the undulating sea surface cause frequent switching between air and underwater controllers, disrupting control continuity and exacerbating attitude oscillations. Summary of the Invention

[0004] The technical problem to be solved by this application is that when current cross-medium unmanned aerial vehicles cross the water-air interface, the physical properties of the medium change abruptly, causing buoyancy, fluid damping, and added mass to change drastically in an instant, which can easily lead to attitude instability, fuselage shaking, or even capsizing. At the same time, repeated high-altitude oscillations near the undulating sea surface cause frequent switching of the air / underwater controller, disrupting control continuity and exacerbating attitude oscillations.

[0005] To address the aforementioned issues, this application provides a tandem dual-rotor unmanned aerial vehicle and its motion control method.

[0006] In a first aspect, the present invention discloses a motion control method for a tandem dual-rotor unmanned aerial vehicle, comprising, The drone's flight altitude is obtained and compared with the preset first altitude threshold and second altitude threshold. The drone's current flight status and the calculation medium immersion ratio coefficient are detected. The drone's thrusters are switched periodically based on the flight status. The first altitude threshold is higher than the second altitude threshold. The system collects real-time status data and desired trajectory data of the UAV, constructs LOS and NTSTSM models, inputs real-time status data, acquires desired trajectory data, and calculates UAV thruster control data using medium immersion ratio analysis. The UAV thrusters operate based on the UAV thruster control data. Status data includes position data, attitude data, velocity data, and disturbance data.

[0007] Preferably, the preceding steps include: Construct a coordinate system, build a model of the UAV on the coordinate system, and label the UAV's state data and thruster control data. The state data includes position data, attitude data, velocity data, and disturbance data. The thruster control data includes thruster speed and thruster tilt angle.

[0008] Preferably, the drone's flight altitude is obtained and compared with preset first and second altitude thresholds. The current flight status of the drone and the immersion ratio of the computing medium are detected. The drone's thrusters are switched periodically based on the flight status. When the first altitude threshold is higher than the second altitude threshold, the specific steps include: A first altitude threshold and a second altitude threshold are preset. If the first altitude threshold is higher than the second altitude threshold, the flight altitude of the drone is acquired in real time, and the flight altitude is compared with the preset first altitude threshold and second altitude threshold to calculate the medium immersion ratio coefficient of the drone. When the drone's flight altitude exceeds the first altitude threshold and the drone is fully in the air, the drone's air thrusters are activated. When the drone's flight altitude is less than or equal to the first altitude threshold and greater than or equal to the second altitude threshold, the drone is partially submerged in water. The control data of the aerial thruster and the control data of the underwater thruster are allocated based on the medium immersion ratio coefficient. When the drone's flight altitude is below the second altitude threshold, the drone is completely underwater, and the drone's underwater thrusters are activated.

[0009] Preferably, real-time status data of the UAV is collected, a LOS guidance law model and an NTSTSM model are constructed, and real-time status data and medium immersion ratio coefficients are input to analyze and calculate the UAV thruster control data. The UAV thruster operates based on the UAV thruster control data, specifically including the following steps: Collect real-time status data of UAVs, filter and fuse the real-time status data to obtain stable status estimates, and calculate the immersion ratio coefficient; Acquire the desired trajectory data of the UAV, construct the LOS guidance law model, input real-time state data and desired trajectory data to calculate the trajectory error, and adjust the desired position data, desired attitude data and desired velocity data based on the trajectory error. The desired trajectory data includes attitude data, position data and velocity data. The updated drone model and its parameters are obtained by weighting the parameters of the drone model based on the medium immersion ratio coefficient. Construct an NTSTSM model, input updated desired trajectory data and real-time status data to calculate tracking error, and generate UAV control torque and total control force based on tracking error and updated UAV model parameters; Based on the medium immersion ratio coefficient, the UAV control torque and total control force are distributed to the air thruster control data and the underwater thruster control data, and the air thruster and underwater thruster are driven to operate based on the air thruster control data and the underwater thruster control data.

[0010] Preferably, the process involves acquiring the desired trajectory data of the UAV, constructing a LOS guidance law model, calculating the trajectory error by inputting real-time state data and desired trajectory data, and adjusting the desired attitude data and desired velocity data based on the trajectory error. Specifically, this includes the following steps: Capture the expected trajectory data of the UAV from the mission parameters of the UAV, and collect real-time status data, including real-time position data, real-time attitude data, real-time speed data, and real-time disturbance data. The trajectory error is obtained by calculating the three-axis tracking error between the real-time position data and the desired position data. The yaw angle, pitch angle, and forward velocity values ​​of the UAV are updated based on the trajectory error to obtain the desired velocity and attitude data.

[0011] Preferably, an NTSTSM model is constructed, the updated desired trajectory data and real-time state data are input to calculate the tracking error, and the UAV control torque and total control force are generated based on the tracking error and the updated UAV model parameters. Specifically, this includes the following steps: Obtain the updated expected trajectory data, real-time status data, medium immersion ratio coefficient, and updated UAV model parameters, and construct the NTSTSM model. The tracking error is calculated by comparing the expected velocity data with the real-time velocity data, and the tracking error is calculated by comparing the expected attitude data with the real-time attitude data, thus obtaining the velocity tracking error and the attitude tracking error. An integral sliding surface is constructed based on the velocity tracking error and the medium immersion ratio coefficient, and a linear sliding surface is constructed based on the attitude tracking error and the medium immersion ratio coefficient, resulting in a non-singular terminal super-helical sliding surface. A continuous superspiral algorithm is used to generate a continuous control law, and the control torque and total control force of the UAV are calculated based on the non-singular terminal superspiral sliding surface and disturbance data.

[0012] Preferably, the drone is fully airborne when its aerial thrusters are activated, specifically including the following steps: When the drone is completely in the air, it collects position data, attitude data, speed data, flight altitude data and wind field information in real time, and obtains stable aerial motion state data after filtering and fusion. The desired yaw angle in the horizontal plane and the desired pitch angle in the vertical plane are calculated based on the deviation between the preset trajectory and the real-time position, and continuous desired attitude data and desired forward velocity data are generated by combining them with the preset reference speed. Construct an NTSTSM controller, calculate velocity and attitude tracking errors, build an integral sliding surface for the velocity loop and a linear sliding surface for the attitude loop, and use a super-spiral approach law to output smooth control torque, outputting total tension and triaxial torque; The thrust and torque are calculated into rotor speed and tilt angle data for the propeller, which then drives the aerial propeller.

[0013] Preferably, the drone is completely underwater, and the underwater thruster of the drone is activated, specifically including the following steps: When the drone is completely submerged in water, it collects information on depth, attitude, speed and ocean currents in real time to calculate the drone's motion relative to the water flow. Construct a line-of-sight guidance law model, calculate the desired heading on the horizontal plane and the desired pitch angle for submersion and surfacing based on the underwater preset trajectory, real-time position and depth deviation, and maintain a constant desired forward speed, and output underwater desired attitude data and desired speed data. The constructed NTSTSM controller takes underwater desired attitude data and desired velocity data as input, calculates velocity and attitude tracking errors based on underwater rigid body dynamics model, constructs integral velocity sliding surface and linear attitude sliding surface, and uses super-helical control algorithm to output propulsion thrust and three-axis control torque and thrust. The thrust and control torque are distributed to the underwater thruster speed and the deflection angles of the four control surfaces of the underwater thruster's X-shaped rudder, driving the underwater thruster to operate.

[0014] Secondly, the present invention discloses a tandem dual-rotor unmanned aerial vehicle (UAV), characterized in that it includes the motion control method for a tandem dual-rotor UAV as described in any one of claims 1-8.

[0015] Preferably, it includes two air thrusters, an underwater thruster, and a fuselage. The fuselage has a spindle-shaped shape. The underwater thruster is installed at one end of the fuselage, and the air thrusters are set on the fuselage. The two air thrusters are arranged in a longitudinal row along the direction of the fuselage.

[0016] The technical solution provided in this application has the following advantages compared with the prior art: This application provides a tandem dual-rotor unmanned aerial vehicle (UAV) and its motion control method. The method involves acquiring the UAV's flight altitude and comparing it with preset first and second altitude thresholds to accurately identify the UAV's flight state and calculate the medium immersion ratio coefficient in real time. Simultaneously, a timed constraint mechanism is used to achieve smooth switching between the thrusters and the controller. Real-time state data such as the UAV's position, attitude, speed, and disturbances are collected, and a LOS guidance law model and an NTSTSM control law model are constructed based on the desired trajectory data. Smooth calculations are performed using the immersion ratio coefficient as the core, outputting continuous and stable thruster control data. This ensures that the force and control commands of the UAV do not experience abrupt changes during water-air cross-medium movement, achieving a smooth transition of buoyancy, fluid damping, and added mass. This effectively avoids the risk of fuselage shaking and capsizing caused by sudden force changes during cross-medium movement. Furthermore, the timed switching constraint mechanism eliminates frequent switching between the air and underwater controllers caused by sea surface fluctuations, ensuring control continuity and attitude stability. The combination of LOS guidance and NTSTSM control enables high-precision trajectory tracking and disturbance suppression, significantly improving the stability, robustness, and operational safety of the UAV during water-air cross-medium movement.

[0017] The tandem dual-rotor UAV mentioned above adopts a streamlined, spindle-shaped fuselage with two aerial thrusters arranged in a tandem along the fuselage's longitudinal direction and an underwater thruster located at the tail. The UAV operates using the cross-medium motion control method described in the first aspect above, which can automatically identify the water-air state based on flight altitude and immersion ratio coefficient, achieving seamless switching between aerial and underwater navigation without interference. Combined with the adopted cross-medium control method, it can maintain stable attitude, without shaking or capsizing, when crossing the water-air interface, while avoiding frequent controller switching, thus improving the overall reliability, continuity, and environmental adaptability of cross-medium navigation. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0019] 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, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a motion control method for a tandem dual-rotor unmanned aerial vehicle provided in this application; Figure 2 A flowchart illustrating step S1 of the motion control method for a tandem dual-rotor unmanned aerial vehicle provided in this application; Figure 3 A flowchart illustrating step S2 of the motion control method for a tandem dual-rotor unmanned aerial vehicle provided in this application; Figure 4 A flowchart illustrating step S22 of the motion control method for a tandem dual-rotor unmanned aerial vehicle provided in this application; Figure 5 A flowchart illustrating step S24 of the motion control method for a tandem dual-rotor unmanned aerial vehicle provided in this application; Figure 6 A cross-media switching control block diagram of a motion control method for a tandem dual-rotor UAV provided in this application; Figure 7 A cross-media switching control signal flowchart for a motion control method for a tandem dual-rotor UAV provided in this application; Figure 8 A schematic diagram of the LOS guidance law for a motion control method of a tandem dual-rotor UAV provided in this application; Figure 9 A schematic diagram of the aerial propulsion system of a tandem dual-rotor unmanned aerial vehicle provided in this application; Figure 10 This application provides a structural diagram of an underwater propulsion system for a tandem dual-rotor unmanned aerial vehicle. Figure 11 A schematic diagram of the aerial flight mode of a tandem dual-rotor UAV provided in this application; Figure 12 This application provides a schematic diagram of the underwater navigation mode of a tandem dual-rotor unmanned aerial vehicle (UAV). Figure 13 A schematic diagram of the water entry and exit motion mode of a tandem dual-rotor UAV provided in this application; Figure 14 A coordinate system definition diagram for a tandem dual-rotor UAV provided in this application; Figure 15 A comparative chart of continuous cross-domain trajectory tracking performance of a tandem dual-rotor UAV provided in this application; Figure 16 This application provides a schematic diagram of crosswind disturbance during a continuous cross-domain process for a tandem dual-rotor unmanned aerial vehicle (UAV). Figure 17 This application provides a diagram illustrating the water flow changes during a continuous cross-domain process of a tandem dual-rotor unmanned aerial vehicle (UAV). Figure 18 This application provides a position change diagram of a tandem dual-rotor UAV during underwater trajectory tracking; Figure 19 A schematic diagram illustrating the positional error of a tandem dual-rotor UAV in underwater trajectory tracking, provided in this application; Figure 20 This application provides an error statistics chart for continuous cross-domain trajectory tracking of a tandem dual-rotor UAV; Figure 21 A schematic diagram of the total root mean square error in the three axes of a tandem dual-rotor UAV provided in this application; Figure 22 This application provides an attitude angle change diagram during continuous cross-medium trajectory tracking of a tandem dual-rotor UAV; Figure 23 This application provides a diagram showing the change in the rotation angle of each tail rudder during continuous cross-medium trajectory tracking of a tandem dual-rotor UAV; Figure 24 This application provides a diagram showing the rotor speed variation during continuous cross-domain trajectory tracking of a tandem dual-rotor UAV. Figure 25 This application provides a diagram showing the change in servo tilt angle during continuous cross-domain trajectory tracking of a tandem dual-rotor UAV. Figure 26 A force variation diagram of a tandem dual-rotor UAV during continuous cross-domain trajectory tracking, provided in this application; Figure 27 This application provides a diagram showing the torque variation on the fuselage of a tandem dual-rotor UAV during continuous cross-domain trajectory tracking. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] Firstly, see Figures 1-8 This invention discloses a motion control method for a tandem dual-rotor unmanned aerial vehicle (UAV), comprising: Step S1: Obtain the drone's flight altitude and compare it with the preset first altitude threshold and second altitude threshold. Detect the drone's current flight status and calculate the medium immersion ratio coefficient. Combine the flight status to periodically switch the drone's thrusters. The first altitude threshold is higher than the second altitude threshold. Step S2: Collect real-time status data and desired trajectory data of the UAV, construct the LOS (Line-of-Sight) model and the NTSTSM (Non-singular Terminal Super-Twisting Sliding Mode) model, input the real-time status data, acquire the desired trajectory data, and analyze the medium immersion ratio coefficient to calculate the UAV thruster control data. The UAV thrusters operate based on the UAV thruster control data. The status data includes position data, attitude data, velocity data, and disturbance data.

[0023] Specifically, in step S1, the real-time flight altitude of the UAV is acquired and compared with preset first and second altitude thresholds, where the first altitude threshold is higher than the second altitude threshold. Based on the altitude range, the UAV is determined to be in one of three states: fully airborne, partially submerged, or fully underwater, and a medium immersion ratio coefficient is calculated based on the altitude. Combining the current flight state with timing constraints, the airborne and underwater thrusters are smoothly switched, prohibiting repeated switching within the timing period. This achieves accurate identification of the UAV's airborne and waterborne medium states, establishes a smooth transition mechanism for buoyancy, damping, and added mass, and avoids frequent switching between the thrusters and controller due to wave fluctuations at the hardware execution level. This ensures continuous and stable operation of the power system, preventing abrupt changes in force during cross-medium transitions, eliminating the risk of attitude instability, fuselage shaking, and capsizing from the source, and preventing oscillations caused by frequent controller switching, thus improving the smoothness and reliability of cross-medium switching.

[0024] Specifically, in step S2, real-time status data such as the UAV's position, attitude, velocity, and disturbances are collected, along with preset desired trajectory data. A LOS (Line of Sight) guidance law model and an NTSTSM (Non-Singular Terminal Super-Helical Sliding Mode) control model are constructed. Real-time status data, desired trajectory data, and medium immersion ratio coefficients are synchronously input into the models for analysis and calculation, resulting in the UAV thruster control data. Ultimately, this drives the aerial and underwater thrusters to operate stably according to control commands. Continuous and smooth desired attitude and velocity commands are generated through the LOS guidance law. High-precision, low-jitter trajectory tracking and attitude control are achieved through the NTSTSM model. The immersion ratio coefficient is used to achieve a smooth weighted transition of the air-water control law, compensating for external disturbances such as wind, waves, and currents. This ensures high trajectory tracking accuracy, smooth control commands, and strong anti-interference capabilities throughout the entire cross-medium process, further consolidating cross-medium motion stability and achieving continuous, reliable, and high-precision air-water cross-medium maneuvering.

[0025] It is understandable that by acquiring the UAV's flight altitude and comparing it with preset first and second altitude thresholds, the flight state of the UAV can be accurately identified, and the medium immersion ratio coefficient can be calculated in real time. At the same time, a timed constraint mechanism can be used to achieve a smooth switch between the thruster and the controller. Then, real-time status data such as the UAV's position, attitude, speed, and disturbances are collected, and a LOS guidance law model and an NTSTSM control law model are constructed by combining the desired trajectory data. The immersion ratio coefficient is used as the core for smooth calculation, and continuous and stable thruster control data is output. This ensures that there are no abrupt changes in the force and control commands of the UAV during the water-air cross-medium process, and achieves a smooth transition of buoyancy, fluid damping, and added mass. This effectively avoids the risk of fuselage shaking and capsizing caused by sudden changes in force during cross-medium movement. At the same time, the timed switching constraint mechanism eliminates the frequent switching between the air and underwater controllers caused by sea surface fluctuations, ensuring control continuity and attitude stability. The combination of LOS guidance and NTSTSM control can achieve high-precision trajectory tracking and disturbance suppression, significantly improving the stability, robustness, and operational safety of the UAV in water-air cross-medium movement.

[0026] Preferably, the preceding steps include: Step S0: Construct a coordinate system, build a model of the UAV on the coordinate system, and label the UAV's state data and thruster control data. The state data includes position data, attitude data, velocity data, and disturbance data. The thruster control data includes thruster rotation speed and thruster tilt angle.

[0027] Specifically, firstly, inertial coordinate systems, body coordinate systems, rotor coordinate systems, and tail rudder coordinate systems are established, and the rules for defining the origin and attitude angles of each coordinate system are clarified. Under a unified coordinate system, a cross-medium kinematic and dynamic model of the UAV is constructed, and state data such as position, attitude, velocity, and disturbances, as well as control data such as thruster speed and thruster tilt angle, are correspondingly labeled in the model, forming a complete mathematical description system for the UAV's state and control quantities. This provides a unified mathematical benchmark and descriptive framework for UAV state perception, trajectory guidance, motion control, and actuator actuation, achieving standardized expression of state and control data, ensuring accurate LOS guidance and NTSTSM control calculations under a consistent coordinate system, and avoiding control calculation errors caused by coordinate confusion.

[0028] The specific steps are as follows: Define the inertial coordinate system, body coordinate system, rotor coordinate system, and tail rudder coordinate system sequentially; clarify the positive and negative rules for the pointing of each axis and attitude angles (roll φ, pitch θ, yaw ψ) to form a unified mathematical benchmark. Define the inertial coordinate system. Its origin is located at a point on sea level. Body coordinate system. The origin is located at the center of gravity of the cross-medium UAV, where The shaft points towards the machine head. The axis points to the bottom of the drone. The axis direction is determined by the right-hand rule. Rotor coordinate system. (in The origins (representing the front and rear rotors respectively) are located at the mounting centers of the two brushless motor tilt shafts. The axis points towards the bow of the drone. The shaft is mounted on the front rotor motor shaft and points towards the bottom of the drone. The axis is determined by the right-hand rule. Tail rudder blade coordinate system. (in The origin of each of the four independent control surfaces (representing the X-shaped tail rudder) is located at the center of the tail propulsion section. The axis points towards the bow of the transmedium UAV. The axis points towards the center of the rudder surface mounting. The axis is determined by the right-hand rule. The projection angle of the cross-medium UAV in the body coordinate system is defined as: roll angle. Machine system and The angle between the vertical plane containing the axis is defined as follows: when the roll angle is positive, the UAV rolls to the right. Pitch angle. Machine system Axis and ground plane The angle between the planes; when the drone pitches up, the pitch angle is positive. Yaw angle. Machine system At ground level projection and The angle between the axes defines the projection onto the axis. When the axis is to the right, the yaw angle is positive. Then, a coordinate transformation relationship is constructed. According to the Newton-Euler equations, this paper considers the transmedium UAV as an ideal homogeneous rigid body, and its kinematic equations can be expressed as: In the formula, relative to the body coordinate system The linear velocity and angular velocity vectors, , Represents relative to the inertial coordinate system Position and attitude vectors Let be the kinematic transformation matrix between the body coordinate system and the inertial coordinate system. For position and attitude, the dynamic equations of a cross-medium UAV can be expressed as follows: ;in, The inertia matrix, The Coriolis force matrix, Here is the damping matrix. For gravity and buoyancy, To control the input, This is a disturbance. In this dynamic model, , This represents the inertia matrix that includes hydrodynamically added mass. The Coriolis force and centripetal matrix of the cross-medium UAV and its added mass can be expressed by the following formula: , In this formula, , , , In the dynamic model, The simplified hydrodynamic damping can be expressed by the following formula: , When this cross-medium UAV navigates underwater, in addition to gravity, it is also subject to buoyancy. The restoring force and torque vector generated by gravity and buoyancy can be used... In an inertial coordinate system, the gravitational and buoyant forces acting on a transmedium UAV can be expressed as: , In the formula For the gravity acting on a cross-medium drone, This refers to the buoyancy force experienced by the cross-medium UAV. State and control data are labeled in the model, where the state data includes position. ,attitude ,speed disturbance The propulsion control data includes the rotor speed in the air. tilt angle underwater thruster speed , rudder deflection angle .

[0029] Step S1 specifically includes the following steps: Step S11: Preset a first altitude threshold and a second altitude threshold. If the first altitude threshold is higher than the second altitude threshold, obtain the flight altitude of the UAV in real time, compare the flight altitude with the preset first altitude threshold and second altitude threshold, and calculate the medium immersion ratio coefficient of the UAV. Step S12: When the drone's flight altitude is greater than the first altitude threshold and the drone is completely in the air, activate the drone's air thrusters; Step S13: When the drone's flight altitude is less than or equal to the first altitude threshold and greater than or equal to the second altitude threshold, the drone is partially submerged in water. The control data of the aerial thruster and the control data of the underwater thruster are allocated based on the medium immersion ratio coefficient. Step S14: When the drone's flight altitude is less than the second altitude threshold, the drone is completely underwater, and the drone's underwater thrusters are activated.

[0030] Specifically, preset a first height threshold h1 and a second height threshold h2, satisfying ; Collect the vertical height in the inertial coordinate system of the UAV in real time , and compare with h1 and h2; Calculate the medium immersion proportionality coefficient ζ according to the height interval, and the formula is as follows: , where, , ζ = 1 means completely in the air, 0 < ζ < 1 means partially immersed in water, and ζ = 0 means completely underwater. Establish a quantitative correspondence between height and medium state to achieve accurate, stable, and non-jumping identification of water-air states, and avoid misjudgment of states caused by small fluctuations in height. When the flight height of the UAV satisfies z > h1, it is determined to be completely in the air. At this time, the immersion proportionality coefficient ζ = 1, the system starts the air thruster, closes the underwater thruster, and the actuator only outputs the control commands of the rotor speed and tilt angle, and enters the pure air control mode. The power system and control law fully match the aerodynamic characteristics to ensure flight efficiency and attitude stability.

[0031] When the flight height satisfies h₂ ≤ z ≤ h₁, it is determined to be partially immersed in water. According to the calculated medium immersion proportionality coefficient ζ, linearly weighted distribution is performed on the control amounts of the air thruster and the underwater thruster. Among them, the weight of the air thruster control data is ζ, and the weight of the underwater thruster control data is 1 ζ, and the control amount distribution formula: , , where is the air control command, , including the front and rear rotor speeds and the rotor tilt angle ; is the underwater control command, , including the tail thruster speed and the X-shaped four-rudder deflection angle vector ; Finally, output the hybrid control command to drive the actuator to make a smooth transition. The complete operation formula for linear weighted distribution in the transition zone: ; is the complete multi-dimensional control command vector finally sent to the actuator, so as to achieve smooth power switching across the medium interface, make the air and underwater thrusters work together, and avoid impacts and vibrations caused by command mutations. When the flight height satisfies z < h₂, it is determined to be completely underwater. At this time, the immersion proportionality coefficient ζ = 0, the system starts the underwater thruster, closes the air thruster, the rotor stops working and folds up, and only the tail thruster and the X-rudder provide thrust and torque.

[0032] As an embodiment, when 0 < In the submerged transition zone (<1), attitude stability is the highest priority. When the vertical height error exceeds a set threshold, the lift weight ζ of the air rotor is increased to suppress sinking. When ocean current disturbance exceeds a threshold, the lift weight of the underwater thruster and X-rudder is increased. ζ counteracts the impact of water flow; when ζ=1 in the air, the underwater thruster is locked and in standby mode; when ζ=0 underwater, the air rotor folds up and stops, and only the underwater power system operates.

[0033] As one example, in the transition mode, the two power systems are redundant. If the air thruster fails, the system forces ζ=0 and immediately switches to pure underwater navigation mode. If the underwater thruster fails, the system forces ζ=1 and quickly lifts the altitude to leave the water surface, relying solely on the air rotor for flight.

[0034] Step S2 specifically includes the following steps: Step S21: Collect real-time status data of the UAV, filter and fuse the real-time status data to obtain a stable status estimate, and calculate the immersion ratio coefficient; Step S22: Obtain the desired trajectory data of the UAV, construct the LOS guidance law model, input the real-time state data and the desired trajectory data to calculate the trajectory error, and adjust the desired position data, desired attitude data and desired velocity data based on the trajectory error. The desired trajectory data includes attitude data, position data and velocity data. Step S23: Weight the parameters of the UAV model based on the medium immersion ratio coefficient to obtain the updated UAV model and the updated UAV model parameters; Step S24: Construct the NTSTSM model, input the updated desired trajectory data and real-time status data to calculate the tracking error, and generate the UAV control torque and total control force based on the tracking error and the updated UAV model parameters; Step S25: Distribute the UAV control torque and total control force to the air thruster control data and underwater thruster control data according to the medium immersion ratio coefficient, and drive the air thruster and underwater thruster to operate based on the air thruster control data and underwater thruster control data.

[0035] Specifically, real-time status data of the UAV is collected, including position (x, y, z) and attitude. The raw data, including θ, ψ, linear velocities u, v, w, angular velocities p, q, r, and disturbance data, are processed using a filtering and state fusion algorithm to obtain stable state estimates. , According to altitude The medium immersion ratio coefficient is calculated using the first and second altitude thresholds to obtain high-precision, low-noise state feedback, providing reliable input for guidance and control, and determining the current water-air mixing degree. Simultaneously, sensor noise and transient disturbances are eliminated, ensuring stable and reliable state estimation and guaranteeing accurate subsequent guidance and control calculations. The desired trajectory data of the UAV, including its desired position, is acquired. , , Expected posture , , Expected speed , , Construct a LOS line-of-sight guidance law model and calculate the trajectory tracking error: , , ; Generate desired attitude and desired velocity based on the error: ; ; The updated desired position, attitude, and velocity are output, generating continuous and smooth guidance commands to guide the UAV along a predetermined trajectory. Based on the medium immersion ratio coefficient ζ, the aerial and underwater dynamic parameters are linearly weighted and fused to obtain the updated UAV model, where the inertial matrix is ​​weighted as follows: Damping matrix weighting: Similarly, by weighting the Coriolis force matrix and the restoring force matrix, we obtain the complete weighted dynamic model: This allows the UAV model to continuously change with the degree of immersion, adapting to the real stress characteristics of water-air cross-medium. A non-singular terminal superspiral sliding mode control (NTSTSM) model is constructed, and the velocity and attitude tracking errors are calculated. , , , , , Construct a sliding surface, including a velocity loop and an attitude loop, wherein the velocity loop is... The attitude ring is The superspiral approach law is used to output control torque and total control force: ;in These are disturbance observations used to compensate for wind, wave, current, and model errors, achieving high-precision, low-bounce trajectory and attitude tracking, suppressing disturbances, and ensuring stable control. Based on the medium immersion proportionality coefficient ζ, the total control force and control torque are distributed to the air / underwater thrusters. The formula for allocating the air thruster control data (weight ζ) is as follows: Underwater thruster control data (weight 1) The allocation formula for ζ is: The control input is calculated as the rotor speed in the air. , tilt angle , and underwater thruster speed , rudder deflection angle It drives the actuator to operate, enabling smooth switching and coordinated output of the aerial and underwater power systems, and avoiding sudden changes in actuator movement.

[0036] The specific operation of step S2 is as follows: When the cross-medium UAV is in a water-air transition state partially submerged in water, the system first collects the three-dimensional position, attitude, angular velocity, linear velocity, wave height, and environmental disturbance information of the UAV in real time through the measurement unit. After filtering and data fusion, a stable state estimate is obtained, and the transition mode is determined based on the altitude range. At the same time, the submersion ratio coefficient is calculated to achieve a linear and smooth transition of force and control parameters. Subsequently, the LOS line-of-sight guidance law calculates the horizontal tracking error and vertical error based on the deviation between the measured position and the desired trajectory. Combined with the submersion ratio coefficient, the air and underwater guidance gains are weighted to output a continuous and smooth desired attitude angle and desired velocity, avoiding abrupt changes in guidance commands. Next, the system performs weighted fusion of the air and underwater dynamic models based on the submersion ratio coefficient, and performs linear interpolation on the inertial matrix, Coriolis force matrix, fluid damping matrix, and restoring force / torque including buoyancy, so that the total external force and total torque change continuously during the cross-medium process, eliminating the impact caused by force step. Based on this, the non-singular terminal superspiral sliding mode controller with an extended state observer calculates the tracking error according to the desired state and the measured state, constructs a weighted integral sliding surface and a linear sliding surface, generates continuous control torque using a superspiral reaching law, and estimates and compensates for the total disturbance caused by wind, ocean currents, and model uncertainties in real time through the extended state observer, ensuring stable attitude and high tracking accuracy during the transition process. Subsequently, in the actuator allocation phase of the transition zone, the system distributes control force and torque to the airborne dual-rotor tilting mechanism, underwater thrusters, and X-shaped rudder according to the immersion ratio, so that the rotor output gradually decreases with the immersion ratio, while the underwater actuator output gradually increases, achieving seamless switching of actuator actions. Finally, the system strictly adheres to the minimum switching interval constraint of the controller to prevent frequent controller switching due to altitude oscillations during wave fluctuations, ensuring smooth, continuous, and reliable motion control throughout the entire immersion transition phase.

[0037] This invention acquires multi-source heterogeneous data on airborne status, including position, attitude, velocity, altitude, and wind field. It employs an extended Kalman filter (EKF) to perform data filtering and fusion. In the preprocessing stage, a unified 10ms sampling period and interpolation-aligned timestamps are used to normalize all sensor data into a unified dimension. The position, attitude, and wind field observation equations are decoupled in layers. Through iterative calculations of the EKF prediction step and update step, sensor noise and instantaneous wave disturbances are filtered out, and a stable, oscillating-free airframe state estimate is output, providing reliable state feedback for subsequent LOS guidance and NTSTSM sliding mode control.

[0038] First, a unified 10ms control cycle is set for satellite positioning, IMU inertial unit, altimeter, and wind sensor. Interpolation is used to align the sampling timestamps of different sensors to eliminate timing deviations. Then, position and altitude are uniformly converted to meters (m), attitude is uniformly converted to radians (rad), and velocity and wind speed are uniformly converted to m / s. All sensor quantities are mapped to the same standard dimension system. Finally, position and altitude are filtered separately using the position observation equation; attitude and angular velocity are filtered using inertial attitude decoupling. Wind field information is used as an external disturbance observation value, only used for EKF disturbance term updates, and does not participate in the iteration of body motion state. This achieves hierarchical processing of data in different dimensions and avoids mutual interference between heterogeneous data.

[0039] The extended Kalman filter prediction formula is as follows: ; ; In the formula, f() is the nonlinear dynamic model of the system (the motion equation of the flying submersible). The optimal state after filtering at time step 1; The control input from the previous moment (motor speed, rudder deflection, etc.); This is a priori state, based solely on model deduction without sensor observation. For the nonlinear model f in η^k 1 | k One Jacobian matrix (linearized); The state error covariance at the previous time step; System noise covariance (wind, ocean currents, model error); The prior error covariance represents the uncertainty of the predicted state.

[0040] The extended Kalman filter update formula is as follows: ; ; In the formula, For the observation function h( Jacobian matrix of ) It is the observation noise covariance (GPS, IMU, depth gauge noise). The weighting coefficients balance the reliability of model predictions and sensor observations. For multi-source sensor observations (position, depth, attitude, flow velocity) after unifying dimensions; To use the predicted state to deduce the theoretical observations, Observe the residuals to determine the deviation between the prediction and the measurement. It is the posterior optimal state, i.e., the steady-state estimate after filtering and fusion, which is fed into the LOS guidance and NTSTSM controller.

[0041] Step S22 specifically includes the following steps: Step S221: Capture the expected trajectory data of the UAV from the mission parameters of the UAV, and collect real-time status data, including real-time position data, real-time attitude data, real-time speed data, and real-time disturbance data. Step S222: Calculate the three-axis tracking error between the real-time position data and the desired position data to obtain the trajectory error; Step S223: Update the yaw angle, pitch angle, and forward velocity values ​​of the UAV based on the trajectory error to obtain the desired velocity data and attitude data.

[0042] Specifically, the system first reads and captures the desired trajectory data, including the desired position, desired attitude, and desired speed, from the drone's preset mission parameters; simultaneously, it collects real-time status data of the drone, including real-time position data, through sensors. Real-time attitude data Real-time speed data And real-time perturbation data, where x is the vertical horizontal coordinate, y is the horizontal horizontal coordinate, and z is the vertical coordinate. For roll angle, The pitch angle, Yaw angle Let be the linear velocity, u be the longitudinal forward velocity of the aircraft, v be the lateral lateral velocity of the aircraft, and w be the vertical climbing velocity of the aircraft. Let ω be the angular velocity (rotational speed), p be the roll rate, q be the pitch rate, and r be the yaw rate. Then, the real-time position data is subtracted from the desired position data to calculate the three-axis tracking error, as shown in the following formula: , , ,in, , , x, y, z represent the desired position, and x, y, z represent the real-time position. , , To address trajectory error, the deviation of the UAV's current position from the desired trajectory is quantified, providing a basis for guidance correction. Finally, based on the trajectory error, the desired yaw angle, desired pitch angle, and desired forward velocity are calculated and updated sequentially. Specifically, the desired yaw angle is updated as follows: Update the desired pitch angle: Update expected forward velocity: Output the updated desired pose data ( , , ) and expected speed data ( , , By correcting guidance commands in real time through trajectory errors, the drone can continuously move closer to the desired trajectory.

[0043] The specific operation of step S22 is as follows: First, the desired trajectory given by the mission planning, the real-time position and heading information output by the measurement unit, and the preset reference speed are received. The three-axis tracking error between the current position and the desired trajectory is calculated first. Then, the desired yaw angle is calculated based on the lateral and longitudinal errors in the horizontal plane, and the desired pitch angle is calculated based on the vertical altitude error. Within the transition range, the pitch control gain is linearly weighted according to the immersion scaling factor for air and underwater parameters, ensuring a smooth transition of pitch commands. Simultaneously, the forward desired speed is also weighted according to the immersion scaling factor, achieving a continuous transition between airborne forward speed and underwater constant speed navigation. Finally, continuous and abrupt desired roll angle, pitch angle, yaw angle, and three-axis desired speed are output, providing stable guidance commands for subsequent dynamic control.

[0044] Step S24 specifically includes the following steps: Step S241: Obtain the updated expected trajectory data, real-time status data, medium immersion ratio coefficient, and updated UAV model parameters, and construct the NTSTSM model; Step S242: Calculate the tracking error between the desired velocity data and the real-time velocity data, and calculate the tracking error between the desired attitude data and the real-time attitude data to obtain the velocity tracking error and the attitude tracking error; Step S243: Construct an integral sliding surface based on the velocity tracking error and the medium immersion ratio coefficient, and construct a linear sliding surface based on the attitude tracking error and the medium immersion ratio coefficient to obtain a non-singular terminal super-spiral sliding surface; Step S244: Use a continuous superspiral algorithm to generate a continuous control law, and calculate the UAV control torque and total control force based on the non-singular terminal superspiral sliding surface and disturbance data.

[0045] Specifically, the updated desired trajectory data, real-time status data, medium immersion ratio coefficient ζ, and weighted updated UAV model parameters are obtained. Based on the above data, an NTSTSM model is constructed, forming a unified sliding mode control framework suitable for water-air cross-medium operation. This framework provides the core algorithm structure for velocity and attitude closed-loop control, compatible with aerial, underwater, and transitional dynamic characteristics. Errors are calculated for both velocity and attitude, with the velocity tracking error calculated based on the desired velocity data and real-time velocity data. , , The attitude tracking error is calculated based on the desired attitude data and the real-time attitude data. , , In the formula: u, v, w are real-time velocity data. , , For the desired speed data, , , For real-time attitude data, , , This provides the desired attitude data. By quantifying the deviation between the actual motion and the desired command, it provides a basis for sliding surface construction and control law output. A medium immersion ratio coefficient is introduced. By weighting the sliding mode gain, a sliding surface for smooth water-air transition is constructed, in which an integral sliding surface is used to construct a velocity loop. Linear sliding surface constructs attitude loops. This combination yields a non-singular terminal superspiral sliding mode surface, ensuring continuous and abrupt changes in the sliding mode surface during the cross-medium process. It adaptively adjusts with the immersion state, achieving water-air unification, chatter-free operation, and finite-time convergence. Among these... , These are the integral gains for air and underwater respectively. For speed tracking error; This refers to attitude tracking error; For the attitude error derivative, , Differential gain for air and underwater applications. A continuous superspiral reaching law is employed, combined with perturbation observations. Calculate the control force and control torque: In the formula: For sliding surface, , To control the gain, The total disturbances from wind, ocean currents, and model errors are used to ultimately output the three-axis control torque. , , With overall control , , By generating smooth, continuous, and robust control outputs through the superspiral algorithm, disturbances can be suppressed, chattering can be eliminated, and tracking errors can be quickly corrected.

[0046] The specific operation of step S24 is as follows: First, based on the desired attitude and velocity output by the guidance law and the actual state fed back by the measurement unit, the velocity tracking error and attitude tracking error are calculated. A transition sliding surface is constructed based on the tracking error, where the velocity loop uses an integral sliding surface and the attitude loop uses a linear sliding surface. The sliding surface gain also smoothly transitions between air and water parameters according to the immersion ratio coefficient. Then, a continuous control law is generated using a non-singular terminal superspiral approach law, effectively suppressing the chattering problem caused by traditional sliding mode control. Simultaneously, the extended state observer (ESO) estimates and compensates for total disturbances such as wind disturbance, ocean current, and model error in real time, ultimately outputting smooth three-axis control force and control torque to ensure stable attitude and high trajectory tracking accuracy of the UAV during cross-medium transitions.

[0047] With the drone fully airborne, activate its aerial thrusters, specifically including the following steps: When the drone is completely in the air, it collects position data, attitude data, speed data, flight altitude data and wind field information in real time, and obtains stable aerial motion state data after filtering and fusion. The desired yaw angle in the horizontal plane and the desired pitch angle in the vertical plane are calculated based on the deviation between the preset trajectory and the real-time position, and continuous desired attitude data and desired forward velocity data are generated by combining them with the preset reference speed. Construct an NTSTSM controller, calculate velocity and attitude tracking errors, build an integral sliding surface for the velocity loop and a linear sliding surface for the attitude loop, and use a super-spiral approach law to output smooth control torque, outputting total tension and triaxial torque; The thrust and torque are calculated into rotor speed and tilt angle data for the propeller, which then drives the aerial propeller.

[0048] Specifically, when the UAV is fully airborne, the system first collects the UAV's three-dimensional position data (x, y, z) and roll, pitch, and yaw attitude data in real time through a satellite positioning module, inertial measurement unit, barometric altimeter, and wind field sensor. The raw signals, including θ, ψ, three-axis velocity and angular velocity data u, v, w, p, q, r, flight altitude data, and wind field, wind speed, and wind direction information, are filtered and fused to obtain stable and reliable estimates of the airborne motion state. and Subsequently, the system uses the desired location data from the preset desired trajectory data. , , The three-axis trajectory tracking error is calculated using real-time position data, and the formula is as follows: , , The desired yaw angle is calculated based on the horizontal trajectory error. The desired pitch angle is calculated based on the vertical height error. And combined with the preset reference speed Generate desired forward velocity This outputs continuous and smooth desired attitude and velocity data in the air. Next, the system constructs a Non-Singular Terminal Superhelical Sliding Mode Control (NTSTSM) controller, which calculates the velocity tracking error based on the desired velocity, desired attitude, and real-time state data. , , With attitude tracking error , , Constructing an integral sliding surface for the velocity loop Construct a linear sliding surface sa = ea + ce˙a for the attitude loop. , These are the in-flight integral gains, For speed tracking error, For attitude tracking error, For the attitude error derivative, For the differential gain in the air, a superspiral reaching law is used. Perform calculations. For sliding surface, , To control the gain, The total tensile force is output as a smooth result of wind and model error disturbances. With roll, pitch, and yaw three-axis control torque , , Finally, the total tension and the three-axis control torque are distributed to the actuators using the tension distribution formula. Pitching moment formula and tilt angle distribution relationship , In the formula, , The thrust of the two sets of rotors, For the lever arms from the centers of the two sets of rotors to the pitch axis of the aircraft, , These are the control proportional coefficients (calibration constants) corresponding to roll and yaw, used to calculate the rotational speeds of the front and rear rotors, a+ce˙a. , These are the in-flight integral gains, For speed tracking error, For attitude tracking error, For the attitude error derivative, For the differential gain in the air, a superspiral reaching law is used. Perform calculations. For sliding surface, , To control the gain, The total tensile force is output as a smooth result of wind and model error disturbances. With roll, pitch, and yaw three-axis control torque , , Finally, the total tension and the three-axis control torque are distributed to the actuators using the tension distribution formula. Pitching moment formula and tilt angle distribution relationship , In the formula, , The thrust of the two sets of rotors, For the lever arms from the centers of the two sets of rotors to the pitch axis of the aircraft, , These are the control proportional coefficients (calibration constants) corresponding to roll and yaw, used to calculate the rotational speeds of the front and rear rotors, a+ce˙a. , These are the in-flight integral gains, For speed tracking error, For attitude tracking error, For the attitude error derivative, For the differential gain in the air, a superspiral reaching law is used. Perform calculations. For sliding surface, , To control the gain, The total tensile force is output as a smooth result of wind and model error disturbances. With roll, pitch, and yaw three-axis control torque , , Finally, the total tension and the three-axis control torque are distributed to the actuators using the tension distribution formula. Pitching moment formula and tilt angle distribution relationship , In the formula, , The thrust of the two sets of rotors, For the lever arms from the centers of the two sets of rotors to the pitch axis of the aircraft, , These are the control proportional coefficients (calibration constants) corresponding to roll and yaw, used to calculate the rotational speeds of the front and rear rotors. , With tilt angle data , It outputs control commands to the electronic speed controller and tilt servo actuator, directly driving the airborne dual-rotor propulsion system to operate stably, achieving zero tilt angle data. , It outputs control commands to the electronic speed controller and tilt servo actuator, directly driving the airborne dual-rotor propulsion system to operate stably, achieving zero tilt angle data. , It outputs control commands to the electronic speed controller and tilt servo drive, directly driving the airborne dual-rotor propulsion system to operate stably, enabling precise maneuvering control of the UAV in a fully airborne state, including vertical take-off and landing, pitch, roll, yaw, and lateral translation.

[0049] With the drone completely underwater, activate its underwater thrusters, which includes the following steps: When the drone is completely submerged in water, it collects information on depth, attitude, speed and ocean currents in real time to calculate the drone's motion relative to the water flow. Construct a line-of-sight guidance law model, calculate the desired heading on the horizontal plane and the desired pitch angle for submersion and surfacing based on the underwater preset trajectory, real-time position and depth deviation, and maintain a constant desired forward speed, and output underwater desired attitude data and desired speed data. The constructed NTSTSM controller takes underwater desired attitude data and desired velocity data as input, calculates velocity and attitude tracking errors based on underwater rigid body dynamics model, constructs integral velocity sliding surface and linear attitude sliding surface, and uses super-helical control algorithm to output propulsion thrust and three-axis control torque and thrust. The thrust and control torque are distributed to the underwater thruster speed and the deflection angles of the four control surfaces of the underwater thruster's X-shaped rudder, driving the underwater thruster to operate.

[0050] Specifically, when the UAV is fully submerged in water and its underwater thrusters are activated, the system first uses depth sensors, inertial measurement units, Doppler logs, and ocean current sensors to collect real-time data on the UAV's diving depth, roll, pitch, and yaw attitude. The data on θ, ψ, three-axis velocities and angular velocities u, v, w, p, q, r, and ocean current velocity are combined with the acquired absolute speed data to determine the motion state of the computer body relative to the water flow. Simultaneously, filtering and fusion processing are performed on the raw sensor data to obtain a stable underwater state estimate. , The system then constructs a Line-of-Sight (LOS) guidance law model, which includes the real-time position and the relative water flow velocity. , depth Desired position corresponding to the underwater preset trajectory , Expected depth The difference is calculated to obtain the position and depth deviation. , , The desired heading angle is calculated based on the horizontal plane deviation. The desired pitch angle for submersible control is obtained by combining depth deviation. And set a constant desired forward speed. Finally, complete underwater desired attitude data is output. , , Compared with expected speed data , , Next, an NTSTSM controller will be built, inputting the desired underwater attitude, desired velocity, and fused real-time state data into the controller, based on the underwater rigid body dynamics equations. Perform calculations to determine the speed tracking error. , , With attitude tracking error , , Constructing an integral velocity sliding surface based on velocity tracking error Constructing a linear attitude sliding surface based on attitude tracking error Then, a superspiral control law is adopted. Complete the calculation and output the underwater axial propulsion thrust. and roll, pitch, and yaw three-axis control torques , , In the formula This includes observed compensation values ​​for external disturbances such as ocean currents and fluid damping. Finally, the propulsion thrust and three-axis control torque are distributed to the actuators, taking into account the relationship between propulsion thrust and rotational speed. The operating speed n of the underwater thruster is calculated, where K, T, 、 To fix the parameters, a pseudo-inverse algorithm for the rudder effect matrix is ​​used. B is the rudder effect matrix, which describes the mapping relationship between the deflection angle of each rudder surface and the output torque. It is the pseudo-inverse matrix of B. The pseudo-inverse algorithm of the rudder effect matrix distributes the control torque to the four control surfaces of the X-shaped rudder, and solves for the deflection angle of each control surface. , , , It also sends all control commands to the corresponding drive modules, driving the underwater thrusters and X-shaped rudders to work together to achieve underwater depth-holding, orientation, stable diving, and various maneuvers for the UAV.

[0051] Secondly, see Figures 9-26 This invention discloses a tandem dual-rotor unmanned aerial vehicle (UAV) driven by a motion control method for tandem dual-rotor UAVs.

[0052] The tandem dual-rotor UAV includes two air thrusters, an underwater thruster, and a fuselage. The fuselage has a spindle-shaped design. The underwater thruster is installed at one end of the fuselage, and the air thrusters are located on the fuselage. The two air thrusters are arranged in a tandem along the direction of the fuselage.

[0053] Specifically, the UAV adopts a streamlined, spindle-shaped fuselage with two aerial thrusters arranged longitudinally along the fuselage and an underwater thruster at the tail. The UAV operates using the cross-medium motion control method described in the first aspect above, which can automatically identify the water-air state based on the flight altitude and immersion ratio coefficient, achieving seamless switching between aerial and underwater navigation without interference. Combined with the cross-medium control method, it can maintain stable attitude, without shaking or capsizing, when crossing the water-air interface, while avoiding frequent controller switching, thus improving the overall reliability, continuity, and environmental adaptability of cross-medium navigation.

[0054] The aerial propulsion system includes a waterproof tilt servo, a tilt support shaft, a tilt motor mount, a waterproof brushless motor, and rotors. Upon receiving a control signal, the waterproof tilt servo drives the tilt motor mount, causing the motor mounted on the mount to tilt around the tilt support shaft axis, thus generating control torque. This torque matches the attitude control requirements of the cross-medium UAV, enabling precise adjustment of roll and yaw. Pitch attitude control is achieved through the differential speed adjustment of two brushless motors, which drive the rotors. The UAV achieves vertical takeoff and landing in confined spaces through two counter-rotating rotors. The underwater propulsion system includes a tail thruster, an X-shaped tail rudder, and small waterproof servos. The X-shaped tail rudder consists of four blades, each independently driven by a small waterproof servo. The tail thruster provides the main thrust for forward and backward navigation. The four blades are driven independently by separate servos, forming an X-shaped layout for more flexible steering, pitch, and roll attitude control.

[0055] The simulation analysis of motion control for cross-medium UAVs is as follows: Considering the complex marine environment that the cross-domain amphibious vehicle will face in actual operations, continuous cross-medium switching motion of the amphibious vehicle near the water surface was simulated under wind, wave, and current conditions. This aims to verify the maneuverability of the designed amphibious vehicle near the water surface and the reliability and stability of the proposed NTSTSM method. Assuming the starting coordinates of this continuous cross-medium motion are (0, 0, 2), the desired trajectory can be represented by the following set of equations: , As discussed earlier, during the movement of a crosswind and ocean current, higher crosswind levels and stronger currents place greater demands on the controller and impose stricter tests on its stability. Therefore, in simulating the cross-medium movement in this section, the underwater stage disturbance is determined to be the effect of a V5 level current velocity. It is also stipulated that upon each surface breakup, the crosswind will be immediately subjected to a continuous V5 level crosswind disturbance. Furthermore, to simulate more realistic sea surface conditions, this study uses generated random wavefronts to distinguish the water-air interface; the wavefront height values ​​are shown in the table below.

[0056] Table 1 Wave surface parameter settings

[0057] Figure 15 shows the continuous cross-medium motion of the cross-domain flying submersible tracking the desired trajectory under the three control methods. It can be seen that all three control methods enable the cross-domain flying submersible to achieve continuous cross-medium motion. This simulation verifies that the designed cross-domain flying submersible possesses cross-medium motion capability and also demonstrates the reliability of the cross-medium switching strategy.

[0058] During this continuous transmedium motion, the effects of crosswinds on the transoceanic submersible and the influence of underwater currents change over time as follows: Figure 16 and Figure 17 As shown, it can be observed that as the transoceanic drone continuously transitions between water and air, it is intermittently affected by crosswinds, resulting in force disturbances in different directions. However, these disturbances cease immediately after each dive. Notably, within a brief interval of approximately 52 to 53 seconds, the crosswind disturbance experienced by the drone abruptly disappears. This phenomenon is primarily due to the surge of waves during this period, which briefly submerges the drone below the water surface, thus eliminating the influence of the crosswinds.

[0059] Figure 18 , Figure 19 and Figure 22 The changes in position, position error, and attitude angle of the cross-domain aero-submarine during its continuous cross-medium motion are shown. Comparison of the control effects of the three methods reveals that under the PID control method, the cross-domain aero-submarine experiences a significant position error in the y-direction when encountering crosswinds after surfacing. The ADRC and NTSTSM control methods demonstrate stronger robustness in handling lateral disturbances, effectively compensating for the inherent shortcomings in y-direction control performance of this type of cross-domain aero-submarine. Further simulation results show that the NTSTSM control method enables the cross-domain aero-submarine to achieve stable trajectory tracking control with minimal error both underwater and in the air. Furthermore, during continuous water-to-air transitions, whether in the surfacing or re-entry phases, the NTSTSM control method exhibits minimal cross-medium switching chatter and achieves relatively faster stable convergence, thus demonstrating smoother switching performance.

[0060] Similarly, the trajectory tracking errors under three control methods during continuous cross-domain processes were also compiled and statistically analyzed, such as... Figure 20 and Figure 21 As shown, compared to the other two methods, the NTSTSM method can achieve the desired trajectory tracking with smaller tracking errors in both the y-axis and z-axis directions. Although it is somewhat inferior to the ADRC method in terms of the main error in the x-axis direction, it still performs well in terms of position across all three axes. The size is still relatively small, and the control performance has good robustness.

[0061] Figure 23 The image shows the rotation angles of each blade of the "X" rudder during continuous cross-domain trajectory tracking by the cross-domain flying submersible. It can be seen that each blade underwent rapid and frequent large-angle adjustments at the moment of exiting or entering the water. Therefore, to achieve rapid and efficient continuous cross-medium movement, both the hardware quality and control precision of the tail rudder must be guaranteed.

[0062] Under the NTSTSM control method, the rotational speed of the air-mounted brushless motor and the servo tilt angle that achieves motor tilting are respectively as follows: Figure 24 and Figure 25 As shown, the rotational speeds of the two motors instantly increase to the required lift with each water-crossing maneuver, maintaining stable flight for a period before shutting down and rapidly re-entering the water. Correspondingly, the two tilt servos also rapidly and drastically adjust during each water-crossing maneuver to achieve attitude stability. Furthermore, during continuous water-crossing maneuvers, the aircraft experiences significant impacts due to the rapid change in medium. For example... Figure 26 As shown, the impact forces experienced by the aircraft in different directions under three control methods are illustrated. It can be seen that the impact force experienced by the aircraft under the PID control method is greater than that under the other two methods. This is because the aircraft's water entry attitude is relatively poor during the cross-medium process, and the transition is not smooth enough. Further comparison... Figure 26 The forces acting on the object in the x and z directions, and Figure 27 As can be seen from the torques in each axis direction, during the continuous water-air crossing process, the force and torque generated at the moment of entering the water are greater than those at the moment of exiting the water. This is because the velocity of motion when entering the water is greater than that when exiting the water.

[0063] This section proposes an NTSTSM control method incorporating ESO and establishes cross-medium switching rules. Based on the LOS guidance method, the tracking control performance of the cross-domain flying submersible in multi-domain scenarios is analyzed. By comparing with PID and ADRC control methods, it is concluded that the NTSTSM control method with ESO can significantly reduce tracking errors and achieve stable tracking of the target trajectory. Furthermore, the control method exhibits good tracking accuracy maintenance capability when facing various disturbances such as crosswinds, ocean currents, and wave fluctuations, and has strong robustness against sudden external disturbances.

[0064] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0065] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0066] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0067] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0068] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0069] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. The illustrative expressions of the above terms in this specification should not be construed as necessarily referring to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0070] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Since these modifications and variations fall within the scope of the claims and their equivalents, this invention also intends to include these modifications and variations.

[0071] The above description describes specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A motion control method for a tandem dual-rotor unmanned aerial vehicle (UAV), characterized in that, include, The drone's flight altitude is obtained and compared with the preset first altitude threshold and second altitude threshold. The drone's current flight status and the calculation medium immersion ratio coefficient are detected. The drone's thrusters are switched periodically based on the flight status. The first altitude threshold is higher than the second altitude threshold. The system collects real-time status data and desired trajectory data of the UAV, constructs LOS guidance law model and NTSTSM model, inputs real-time status data, acquires desired trajectory data, and calculates UAV thruster control data using medium immersion ratio analysis. The UAV thruster operates based on the UAV thruster control data. Status data includes position data, attitude data, velocity data, and disturbance data.

2. The method according to claim 1, characterized in that, The previous steps included: Construct a coordinate system, build a model of the UAV on the coordinate system, and label the UAV's state data and thruster control data. The state data includes position data, attitude data, velocity data, and disturbance data. The thruster control data includes thruster speed and thruster tilt angle.

3. The method according to claim 1, characterized in that, The drone's flight altitude is obtained and compared with preset first and second altitude thresholds. The drone's current flight status and the calculation medium immersion ratio are detected. Based on the flight status, the drone's thrusters are switched periodically. If the first altitude threshold is higher than the second altitude threshold, the specific steps include: A first altitude threshold and a second altitude threshold are preset. If the first altitude threshold is higher than the second altitude threshold, the flight altitude of the drone is acquired in real time, and the flight altitude is compared with the preset first altitude threshold and second altitude threshold to calculate the medium immersion ratio coefficient of the drone. When the drone's flight altitude exceeds the first altitude threshold and the drone is fully in the air, the drone's air thrusters are activated. When the drone's flight altitude is less than or equal to the first altitude threshold and greater than or equal to the second altitude threshold, the drone is partially submerged in water, and the control data of the aerial thruster and the underwater thruster are allocated based on the medium immersion ratio coefficient. When the drone's flight altitude is below the second altitude threshold, the drone is completely underwater, and the drone's underwater thrusters are activated.

4. The method according to claim 1, characterized in that, Real-time status data of the UAV is collected, and a LOS guidance law model and an NTSTSM model are constructed. Real-time status data and medium immersion ratio coefficients are input to analyze and calculate the UAV thruster control data. The UAV thruster operates based on this control data. The specific steps include: Collect real-time status data of UAVs, filter and fuse the real-time status data to obtain stable status estimates, and calculate the immersion ratio coefficient; Acquire the desired trajectory data of the UAV, construct the LOS guidance law model, input real-time state data and desired trajectory data to calculate the trajectory error, and adjust the desired position data, desired attitude data and desired velocity data based on the trajectory error. The desired trajectory data includes attitude data, position data and velocity data. The updated drone model and its parameters are obtained by weighting the parameters of the drone model based on the medium immersion ratio coefficient. Construct an NTSTSM model, input updated desired trajectory data and real-time status data to calculate tracking error, and generate UAV control torque and total control force based on tracking error and updated UAV model parameters; Based on the medium immersion ratio coefficient, the UAV control torque and total control force are distributed to the air thruster control data and the underwater thruster control data, and the air thruster and underwater thruster are driven to operate based on the air thruster control data and the underwater thruster control data.

5. The method according to claim 1, characterized in that, The process of acquiring the desired trajectory data of the UAV, constructing a LOS guidance law model, calculating the trajectory error by inputting real-time state data and desired trajectory data, and adjusting the desired attitude and velocity data based on the trajectory error includes the following steps: Capture the expected trajectory data of the UAV from the mission parameters of the UAV, and collect real-time status data, including real-time position data, real-time attitude data, real-time speed data, and real-time disturbance data. The trajectory error is obtained by calculating the three-axis tracking error between the real-time position data and the desired position data. The yaw angle, pitch angle, and forward velocity values ​​of the UAV are updated based on the trajectory error to obtain the desired velocity and attitude data.

6. The method according to claim 1, characterized in that, Constructing the NTSTSM model, inputting the updated desired trajectory data and real-time state data to calculate the tracking error, and generating the UAV control torque and total control force based on the tracking error and the updated UAV model parameters, specifically includes the following steps: Obtain updated expected trajectory data, real-time status data, medium immersion ratio coefficient, and updated UAV model parameters, and construct the NTSTSM model; The tracking error is calculated by comparing the expected velocity data with the real-time velocity data, and the tracking error is calculated by comparing the expected attitude data with the real-time attitude data, thus obtaining the velocity tracking error and the attitude tracking error. An integral sliding surface is constructed based on the velocity tracking error and the medium immersion ratio coefficient, and a linear sliding surface is constructed based on the attitude tracking error and the medium immersion ratio coefficient, resulting in a non-singular terminal super-helical sliding surface. A continuous superspiral algorithm is used to generate a continuous control law, and the control torque and total control force of the UAV are calculated based on the non-singular terminal superspiral sliding surface and disturbance data.

7. The method according to claim 1, characterized in that, With the drone fully airborne, activate its aerial thrusters, specifically including the following steps: When the drone is completely in the air, it collects position data, attitude data, speed data, flight altitude data and wind field information in real time, and obtains stable aerial motion state data after filtering and fusion. The desired yaw angle in the horizontal plane and the desired pitch angle in the vertical plane are calculated based on the deviation between the preset trajectory and the real-time position, and continuous desired attitude data and desired forward velocity data are generated by combining them with the preset reference speed. Construct an NTSTSM controller, calculate velocity and attitude tracking errors, build an integral sliding surface for the velocity loop and a linear sliding surface for the attitude loop, and use a super-spiral approach law to output smooth control torque, outputting total tension and triaxial torque; The thrust and torque are calculated into rotor speed and tilt angle data for the propeller, which then drives the aerial propeller.

8. The method according to claim 1, characterized in that, With the drone completely underwater, activate its underwater thrusters, which includes the following steps: When the drone is completely submerged in water, it collects information on depth, attitude, speed and ocean currents in real time to calculate the drone's motion relative to the water flow. Construct a line-of-sight guidance law model, calculate the desired heading on the horizontal plane and the desired pitch angle for submersion and surfacing based on the underwater preset trajectory, real-time position and depth deviation, and maintain a constant desired forward speed, and output underwater desired attitude data and desired speed data. The constructed NTSTSM controller takes underwater desired attitude data and desired velocity data as input, calculates velocity and attitude tracking errors based on underwater rigid body dynamics model, constructs integral velocity sliding surface and linear attitude sliding surface, and uses super-helical control algorithm to output propulsion thrust and three-axis control torque and thrust. The thrust and control torque are distributed to the underwater thruster speed and the deflection angles of the four control surfaces of the underwater thruster's X-shaped rudder, driving the underwater thruster to operate.

9. A tandem dual-rotor unmanned aerial vehicle, characterized in that, The motion control method for a tandem dual-rotor unmanned aerial vehicle as described in any one of claims 1-8 above.

10. The tandem dual-rotor UAV according to claim 1, characterized in that, It includes two air thrusters, an underwater thruster, and a fuselage. The fuselage has a spindle-shaped shape. The underwater thruster is installed at one end of the fuselage, and the air thrusters are set on the fuselage. The two air thrusters are arranged in a longitudinal row along the direction of the fuselage.

Images