A control system of a power multiplex type amphibious unmanned aerial vehicle

Abstract

This invention belongs to the field of unmanned aerial vehicle (UAV) control technology and relates to a control system for a power-reusable amphibious UAV. The system includes a bottom-level motion controller consisting of an airborne flight controller, a ground motion controller, and a land-to-air mode switching controller. The airborne flight controller includes a dual-closed-loop cascaded PID controller and an MPC controller. The MPC controller calculates the optimal control signal as its output signal based on the target attitude command provided by the PID controller. The ground motion controller includes a dual-closed-loop PID controller that improves the error between the real-time motion state and the target command. The land-to-air mode switching controller, upon receiving a "land-to-air" switching command or when the real-time flight altitude obtained using barometer, GPS, and IMU sensors reaches a threshold, adjusts the rotor tilt angle to complete the land-to-air or air-to-land mode conversion and switches the bottom-level controller to either the airborne flight controller or the ground motion controller.

Description

Technical Field

[0001] This invention belongs to the field of amphibious unmanned aerial vehicle (UAV) control technology, specifically relating to a control system for a power-reusable amphibious UAV. Background Technology

[0002] Intelligent unmanned platforms, designed entirely around mission objectives and eliminating concerns about human casualties, are increasingly replacing humans in fields such as disaster relief and deep space exploration due to their low cost and flexible configuration. However, individual drones and unmanned vehicles face multiple challenges when dealing with complex terrain environments, exhibiting limitations such as restricted terrain adaptability and narrow environmental perception range, making it difficult to meet the high-timeliness and high-precision monitoring and perception requirements in unstructured terrain.

[0003] Amphibious unmanned aerial vehicles (UAVs) combine the advantages of UAV aerial flight and unmanned vehicle land mobility through heterogeneous integration and multimodal switching technologies. They can perform tasks in both air and land modes and achieve cross-domain mode switching. Leveraging their multimodal advantages, they can complete diverse tasks such as three-dimensional and unstructured terrain reconnaissance, target identification, and material delivery, providing a more flexible and efficient solution for modern monitoring tasks. However, existing amphibious UAVs, while integrating amphibious functions, inevitably suffer from increased system complexity and overall weight, making it difficult to meet the requirements of high-precision, large-scale, and long-endurance monitoring tasks.

[0004] Therefore, this invention addresses the problems of complex structure and high energy consumption faced by existing amphibious UAVs. Guided by the concept of integrated and lightweight configuration design, it proposes a control system for reusable amphibious UAVs to improve their terrain adaptability, maneuverability, and endurance. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a control system for a power-reusable amphibious unmanned aerial vehicle that can effectively control cross-domain land-air movement and mode switching.

[0006] To achieve the objectives of this invention, the following technical solutions will be adopted.

[0007] A control system for a power-reusable amphibious unmanned aerial vehicle (UAV) includes: a remote control module for applying control commands to a lower-level control module, planning paths, monitoring motion status in real time, and recording status data; a communication module for sending control signals from the remote control module to the lower-level control module and feeding back status information measured by the lower-level control module to the remote control module; and a drive module for driving the rotor motor to rotate and controlling its speed via an electronic speed controller or driving the rotor to tilt via a servo motor, based on the control signals input from the lower-level control module; wherein:

[0008] The underlying control module includes a motion controller, a pressure gauge, an IMU, and a GPS. The motion controller comprises an airborne flight controller, a ground motion controller, and a land-air mode switching controller, wherein:

[0009] The airborne flight controller includes a dual-closed-loop cascaded outer-loop PID position controller and an inner-loop MPC attitude controller, a control distributor, and actuators. The MPC attitude controller compares the target attitude command provided by the PID position controller with the output of the prediction model collected by the feedback correction loop, and performs online rolling calculations using the optimal control law, i.e., solving the preset cost function to obtain the optimal control sequence in the control time domain. The optimal control quantity in the control sequence at the current moment is used as the output signal of the MPC attitude controller. The output signals of the PID position controller and the MPC attitude controller are respectively input to the control distributor, and after distribution, the actuators are driven to perform actions to achieve airborne flight motion control.

[0010] The ground motion controller includes a dual-loop PID controller, ground wheel and rotor speed conversion and actuator. The speed and yaw rate are used as inputs to the dual-loop PID controller. The dual-loop PID controller is used to improve the error between the real-time motion state and the target command. After the ground wheel and rotor speed is converted, the actuator is driven to perform actions, apply force and torque to the fuselage to achieve forward and turning control.

[0011] The land-to-air mode switching controller includes a mode switching judgment and a servo controller. When the mode switching judgment receives a "land-to-air" mode switching command, it adjusts the rotor tilt angle through the servo controller to complete the land-to-air mode conversion and switches the ground motion controller to the air flight controller. When the real-time flight altitude obtained by the barometer, GPS and IMU sensors reaches the threshold, it adjusts the rotor tilt angle through the servo controller to complete the "air-to-land" mode conversion and switches the air flight controller to the ground motion controller.

[0012] Furthermore, the target attitude command is obtained in the following manner:

[0013] Based on the constructed nonlinear dynamics model of the amphibious rotorcraft UAV in flight:

[0014] (1);

[0015] In the formula: It is a position vector; The velocity vector in the body coordinate system; The velocity vector in the inertial coordinate system; The coordinate rotation matrix; This is the vector of the body's rotational angular velocity; To combine external forces; This is the attitude angle vector; This is the attitude transformation matrix; It is the inertial tensor; The resultant torque;

[0016] It can be seen that the essence of the attitude change of the UAV is to eliminate the horizontal position error, that is:

[0017] (2);

[0018] In the formula: This refers to the horizontal position error, including errors in the x-axis and y-axis directions; This refers to the actual horizontal position. The target horizontal position;

[0019] For a horizontal position controller using PID control, the control law satisfies:

[0020] (3);

[0021] In the formula: , , These are the gain parameters for the proportional, integral, and derivative components of the horizontal position PID controller, respectively.

[0022] To achieve equation (3), the horizontal acceleration The following conditions must be met:

[0023] (4);

[0024] Therefore, we can conclude that:

[0025] (5);

[0026] In the formula: This is the attitude angle command; This is the angle transformation matrix;

[0027] Receive target attitude command This provides input commands to the MPC attitude controller in the attitude inner loop.

[0028] Furthermore, the working principle of the dual-closed-loop cascaded MPC-PID controller can be analyzed as a time-series recursive optimization process: at the current sampling time k, based on the embedded prediction equation and prediction model, the historical flight dataset and the executed pose control sequence are fused to construct k+1 to k+N... p The system state trajectory prediction at time N is obtained by optimizing the extreme value of the cost function and considering the boundary conditions of the motor's amplitude limit and state space constraints. CThe optimal pose control solution sequence of step control is obtained, and the first quantity of the optimal solution sequence is injected into the control system as the execution quantity. When the system progresses to time k+1, the state observations are re-sampled for state estimation, thereby refreshing the predicted trajectory and solving the UAV pose optimization problem in the finite time domain again, forming a rolling optimization closed-loop architecture.

[0029] Furthermore, the specific process of the time-series recursive optimization process is as follows:

[0030] Based on the established nonlinear dynamics model of the amphibious rotorcraft UAV in flight, a state-space equation is used to describe its linear discrete state-space model, which is then derived as follows:

[0031] (6);

[0032] In the formula: For the pose and state variables of the amphibious unmanned aerial vehicle; For output variables; To control the input amount; External disturbance quantity; A is the state matrix; B1 is the input control matrix; B2 is the disturbance control matrix; C is the output matrix; ignoring external input disturbance, set... The matrix parameters in equation (6) can be determined based on the dynamic parameters of the amphibious rotary-wing UAV.

[0033] By rewriting the linear discrete state-space model in incremental form to reduce the impact of static errors on the system's responsiveness, we have:

[0034] (7);

[0035] In the formula: These are the differences between state variables; To control the difference in scores;

[0036] To simplify the complexity of the control code and adapt it to the hardware computing power and storage space of the underlying flight controller, the attitude controller in the amphibious UAV's airborne flight controller is designed as an MPC controller; therefore, the state variables of the prediction model are selected as the attitude angles and angular velocities of the amphibious UAV, which leads to:

[0037] (8);

[0038] To predict the future flight attitude behavior of amphibious UAVs, a reasonable prediction equation needs to be constructed to optimize the future control trajectory. The future system state variables and future control variables are treated as adjustable variables. Within a pre-selected prediction time domain, where k represents the current sampling time, the future control trajectory and state variables are described as follows:

[0039] (9);

[0040] In the formula: To control the step size; To predict the step size;

[0041] The output of the pre-prediction is represented as:

[0042] (10);

[0043] Based on equation (9), the future predicted output is represented in vector form as follows:

[0044] (11);

[0045] Separately , Then the above formula simplifies to:

[0046] (12);

[0047] By defining the boundaries between control inputs and state variables, a dynamic balance between control performance and physical constraints is achieved. When designing the MPC attitude controller, only the control variables are considered. and its rate of change Constraints; assigning the constraint on the rate of change of input to a single variable. During the sampling time, the following vector relationship is satisfied:

[0048] (13);

[0049] In the formula: To constrain the lower limit of the control quantity; The upper limit of the control quantity constraint conditions; the above two limiting parameters need to be determined in conjunction with the selection of motor and propeller;

[0050] The process of optimizing the flight attitude of an amphibious UAV using a dual-closed-loop cascaded MPC-PID controller can essentially be reduced to solving an optimization problem of a preset performance function. Therefore, it is necessary to define a corresponding cost function to quantify the attitude control objective and solve it. The weighted error between the predictive control output and the reference quantity is combined with the weighted control quantity to construct the optimal cost function. Solving for the optimal value of this function yields the optimal control sequence in the control time domain.

[0051] For amphibious unmanned aerial vehicles (UAVs), the attitude control objective focuses on fast, accurate, and stable tracking of commands. Therefore, the cost function of the MPC attitude controller is expressed as:

[0052] (14);

[0053] In the formula: The output weighting factor for the prediction error; To control the weighting factor of the increment, The reference input sequence;

[0054] Therefore, the problem of controlling the flight attitude of amphibious UAVs can be reduced to an optimization problem of solving the following equation:

[0055] (15);

[0056] The optimal control sequence obtained by solving the above problem is used as the output signal of the MPC controller.

[0057] Furthermore, the dual closed-loop PID ground motion controller consists of a velocity loop PID controller that addresses the problem of lag in the motion speed response of amphibious UAVs in ground mode and an angular velocity loop PID controller that improves the problem of lag in the angular velocity response of amphibious UAVs in ground mode.

[0058] Furthermore, the control law of the dual closed-loop PID ground motion controller satisfies:

[0059] (16);

[0060] In the formula: The speed control quantity output by the speed PID controller; , , These are the gain coefficients of the proportional, integral, and derivative components of the speed PID controller, respectively. The velocity error is in the ground mode. The angular velocity control quantity output by the yaw angular velocity PID controller; , , These are the gain coefficients of the proportional, integral, and derivative components of the yaw rate PID controller; This represents the yaw rate error in the ground mode.

[0061] Furthermore, the ground wheel and rotor speeds are converted into the angular velocity control quantity output by the yaw rate PID controller. And the speed control quantity output by the speed PID controller This is converted into control signals to drive the rotor motor.

[0062] Furthermore, the specific process of obtaining the control signal involves: combining the control quantities of the velocity loop and the angular velocity loop. and Further processing revealed that the expected rotational speeds of the ground wheels on both the left and right sides met the following requirements:

[0063] (17);

[0064] In the formula: The expected rotational speed of the left ground wheel; The expected rotational speed of the right ground wheel; B is the diameter of the ground wheel; r is the fuselage semi-wheelbase.

[0065] Assuming that the rotor motor drive signal and the expected rotational speed of the ground wheel satisfy a linear relationship, the drive control signals of the left and right rotor motors can be expressed as follows:

[0066] (18);

[0067] In the formula: This is the motor control signal for the left ground wheel; This is the motor control signal for the right-side ground wheel; This is the rotational speed conversion factor;

[0068] After the drive control signals from the two ground wheels are transmitted to the motor controller, the actuator motor responds to the signals, driving the motor to rotate and generate forward thrust; based on the response characteristics of the brushless DC motor:

[0069] (19);

[0070] In the formula: , These represent the motor speeds in the left and right ground wheels, respectively. The time constant of the motor; This refers to the slope parameter of the motor curve; These are the parameters for the motor curve constants;

[0071] Based on the dynamic characteristics of the motor and the ground friction coefficient, the relationship between the thrust and slip torque generated by the rotor motor and the motor speed is expressed as follows:

[0072] (20);

[0073] In the formula: The thrust generated by the left ground wheel; The thrust generated by the right ground wheel; This is the ground wheel thrust coefficient; The slip moment generated by the ground wheel; The difference in thrust generated by the differential speed of the two ground wheels; This refers to the rotor tilt angle.

[0074] Furthermore, the control equations of the land-air mode switching controller are expressed as follows:

[0075] (twenty one);

[0076] In the formula: The control quantity that actually acts on the controlled object; The control quantity output by the flight motion controller; The control quantity output by the ground motion controller; It is a modal identifier, including two states: air mode and land mode.

[0077] Furthermore, the servo controller controls the servo through the control quantity provided by the control channel based on the mode switching judgment.

[0078] Beneficial effects: (1) The proposed power reuse type amphibious unmanned aerial vehicle control system relies on the air flight mechanism to enable the unmanned aerial vehicle to have flight movement function, which can cross terrain obstacles and achieve large-scale rapid monitoring coverage; relying on the ground wheel mechanism, the unmanned aerial vehicle can realize ground movement function, which can extend the endurance time with low energy consumption.

[0079] (2) The air flight controller in the proposed power reuse type amphibious UAV control system adopts the MPC-PID control algorithm. Under the premise of meeting the limitations of the real-time computing power and storage capacity of the flight control hardware, the performance of the underlying attitude controller is optimized based on the rolling optimization mechanism of the MPC algorithm. This avoids problems such as cumbersome parameter tuning and strong reliance on experience, while improving the tracking accuracy and anti-disturbance capability of the attitude command.

[0080] (3) The ground motion controller in the proposed power reuse type amphibious UAV control system generates forward thrust through the tilted rotor, drives the ground driven wheel forward based on the dual closed-loop PID controller, and achieves steering by means of wheel differential speed, thereby realizing the reuse of the power system, effectively avoiding the redundancy of the ground drive system, and realizing the weight reduction of the whole machine.

[0081] (4) The proposed power-reusable amphibious unmanned aerial vehicle control system has the ability to switch modes under appropriate conditions. It can switch between air flight mode and ground motion mode by controlling the tilt of the rotor motor according to the external environment and mission instructions, and at the same time complete the switching of the underlying motion controller to ensure that a series of cross-domain operations can be successfully realized. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of the overall structure of a reusable land-air amphibious unmanned aerial vehicle.

[0083] Figure 2 Front view of the overall structure of a reusable land-air amphibious unmanned aerial vehicle;

[0084] Figure 3Left view of the overall structure of a reusable land-air amphibious unmanned aerial vehicle;

[0085] Figure 4 Top view of the overall structure of a reusable land-air amphibious unmanned aerial vehicle;

[0086] Figure 5 A partially enlarged view of the rotor tilting mechanism of a reusable land-air amphibious unmanned aerial vehicle;

[0087] Figure 6 A schematic diagram showing the rotor tilt direction of a reusable land-air amphibious unmanned aerial vehicle.

[0088] Figure 7 This is a diagram illustrating the overall architecture of a power-reusable amphibious unmanned aerial vehicle (UAV) control system.

[0089] Figure 8 This is an architecture diagram of an airborne flight controller based on MPC-PID dual closed-loop cascade.

[0090] Figure 9 This is an architecture diagram of a ground motion controller based on a dual-closed-loop PID controller.

[0091] Figure 10 A diagram illustrating the "land-to-air" switching process of a reusable amphibious unmanned aerial vehicle (UAV).

[0092] Figure 11 This diagram illustrates the "air-to-land" switching process of a reusable amphibious unmanned aerial vehicle (UAV).

[0093] Figure 12 Architecture diagram of a land-air mode switching controller for a reusable land-air amphibious unmanned aerial vehicle;

[0094] Wherein: 1 is the main fuselage; 2 is the rotor mechanism; 3 is the tilt rotor mechanism; 4 is the ground wheel mechanism; 101 is the carbon fiber fuselage truss; 102 is the carbon fiber arm; 103 is the copper connecting column; 104 is the bolt; 201 is the dual-bladed propeller; 202 is the rotor motor; 203 is the motor mount; 301 is the synchronous pulley; 302 is the synchronous belt; 303 is the tilt servo; 304 is the bearing; 401 is the fiberglass hub; 402 is the carbon fiber rim; 403 is the 3D printed connector. Detailed Implementation

[0095] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0096] I. Overall Design of Powered Reusable Land-Air Amphibious Unmanned Aerial Vehicle

[0097] 1. Power-reusable amphibious unmanned structure configuration scheme

[0098] The aforementioned power-reusable amphibious UAV includes a main fuselage 1, a rotor mechanism 2, a tilt rotor mechanism 3, and a ground wheel mechanism 4. In aerial mode, it employs an H-shaped quadcopter structure, achieving six degrees of freedom (elevation, pitch, roll, and yaw) by controlling the speeds of four sets of motors. In ground mode, the tilted rotor generates forward thrust, driving the ground-driven wheels forward, and steering is achieved through differential speeds between the wheel sets. The overall structural shape of the power-reusable amphibious UAV is as follows: Figure 1 As shown, the overall structural front view is as follows: Figure 2 As shown, the left view of the overall structure is as follows: Figure 3 As shown, the overall structure top view is as follows Figure 4 As shown.

[0099] The main fuselage 1 includes: a carbon fiber fuselage truss 101; a carbon fiber arm 102; a copper connecting column 103; and bolts 104. The rotor mechanism 2 includes: a dual-bladed propeller 201; a rotor motor 202; and a motor mount 203. The tilt rotor mechanism 3 includes: a synchronous pulley 301; a synchronous belt 302; a tilt servo motor 303; and a bearing 304. The ground wheel mechanism 4 includes: a fiberglass hub 401; a carbon fiber rim 402; and 3D-printed connectors 403.

[0100] The two carbon fiber fuselage trusses 101 of the reusable amphibious UAV are connected as a whole by 10 copper connecting columns 103 and 20 bolts 104. Rotor motors 202 and dual-bladed propellers 201 are symmetrically mounted on the through rods of carbon fiber arms 102 on both sides of the carbon fiber fuselage trusses 101. The ground wheels on the left and right sides are connected to the carbon fiber arms 102 and the carbon fiber fuselage trusses 101 respectively via bearings 304. The tilt servo motor 303 is connected to the carbon fiber arms 102 via synchronous pulleys 301 and synchronous belts 302, as detailed below. Figure 5 As shown. The tilt servo 303 maintains the attitude stability of the rotor motor 202 by outputting torque. Simultaneously, it drives the carbon fiber arm 102 to tilt the rotor motor 202 through the transmission between the synchronous pulley 301 and the synchronous belt 302, ensuring that the rotor motor 202 can tilt to the corresponding position in any mode. The tilt direction of the rotor motor 202 is illustrated below. Figure 6 As shown.

[0101] 2. Overall Architecture of the Control System for Powered Reusable Land-Air Amphibious Unmanned Aerial Vehicles

[0102] To enhance the scientific rigor and maintainability of the control system design, this invention employs a modular design approach, dividing the entire control system into four main parts based on functional differences: a low-level control module, a remote control module, a communication module, and a drive module. While maintaining independent functions, each module operates collaboratively to ensure the overall functionality of the control system. The overall architecture of the power-reusable amphibious unmanned aerial vehicle (UAV) control system is as follows: Figure 7 As shown.

[0103] The main functions of the underlying control module are as follows: the flight controller receives external command signals and processes them in real time according to the corresponding control strategy, then sends control signals to the drive module; it processes the collected sensor data and feeds it back to the flight controller; it measures various motion states of the amphibious UAV through a series of sensors and transmits the data to the remote control terminal via the communication module. The main functions of the remote control module are as follows: it applies control commands to the amphibious UAV; it performs path planning for the amphibious UAV; it monitors the motion state of the amphibious UAV in real time and records various status data in the form of a flight log. The main functions of the drive module are as follows: it implements power supply and energy distribution; it receives control signals from the flight controller, drives the rotor motor to rotate to generate lift, and controls the rotor speed; it drives the servo motor to control rotor tilt. The main functions of the communication module are as follows: it receives control signals from the remote control terminal and simultaneously sends information to the flight controller; it feeds back the status information measured by the sensors to the remote control terminal via the flight controller.

[0104] II. Dynamic Modeling of Power-Reusable Amphibious Unmanned Aerial Vehicles

[0105] Based on the aforementioned characteristics of the reusable amphibious unmanned aerial vehicle (UAV) configuration, this invention employs the concept of "fractal deconstruction" to establish dynamic models of the reusable amphibious UAV in both air flight and ground motion modes, targeting the equivalent objects in its two modes.

[0106] For the aforementioned aerial flight modal dynamics model, it can be equivalent to an H-type quadcopter UAV. Assuming that its geometric center coincides with the center of gravity of the aircraft, it is a rigid body. Then, the rigid body nonlinear dynamics model of the power-reusable amphibious UAV in aerial flight modes can be obtained as follows:

[0107] (twenty two);

[0108] In the formula: It is a position vector; The velocity vector in the body coordinate system; The velocity vector in the inertial coordinate system; The coordinate rotation matrix; This is the vector of the body's rotational angular velocity; To combine external forces; This is the attitude angle vector; This is the attitude transformation matrix; It is the inertial tensor; This is the resultant torque.

[0109] To simplify the model and facilitate description, the control signals are separated. Taking rotor speed as the control input, the model is decomposed into four independent control channels. The relationship between rotor speed and the four independent control variables—height, pitch, roll, and yaw—satisfies the following:

[0110] (twenty three);

[0111] In the formula, , , , These are control signals for four channels: elevator, pitch, roll, and yaw. , , , The thrust generated by the four rotor motors; , , , This refers to the motor speed; This refers to the fuselage's semi-wheelbase. This is the propeller thrust coefficient; This is the drag coefficient.

[0112] Based on the above analysis, the nonlinear dynamic model of the power-reusable amphibious UAV in the air flight mode can be expressed as:

[0113] (twenty four);

[0114] For the ground motion modal dynamics model, the following assumptions are also made: the distance from the center of mass to the front and rear axles of the fuselage is equal; and since the air resistance experienced by the fuselage is small when traveling at low speeds, air resistance is ignored.

[0115] The displacement and yaw angle of the reusable land-air amphibious UAV in the ground coordinate system can be summarized as follows:

[0116] (25);

[0117] When a powered, reusable amphibious unmanned aerial vehicle (UAV) moves on the ground, according to Newton's second law, its trajectory along... axis, The force equilibrium equations for the axis and the axis about the center of mass, i.e., the dynamic model under the ground mode, are as follows:

[0118] (26);

[0119] In the formula, m is the mass of the entire machine; Displacement along the x-axis; Displacement along the y-axis; Yaw angle; , The azimuth force acting on the left and right ground wheels; The yaw moment is the force applied. This refers to the fuselage wheelbase; Yaw angular velocity; Let be the moment of inertia about the z-axis.

[0120] III. Design of a Motion Controller for a Power-Reusable Amphibious Land-Air UAV

[0121] 1. Airborne flight controller based on MPC-PID dual closed-loop cascade

[0122] To address the flight motion control problem of powered amphibious unmanned aerial vehicles (UAVs), this invention proposes an MPC-PID dual-closed-loop cascaded aerial flight controller. This controller employs PID control in the outer position loop and MPC control in the inner attitude loop. While meeting the real-time computing power and storage capacity limitations of the flight control hardware, it optimizes the performance of the underlying attitude controller based on the rolling optimization mechanism of the MPC algorithm. The architecture of the MPC-PID-based dual-closed-loop cascaded aerial flight controller is as follows: Figure 8 As shown.

[0123] Based on the constructed nonlinear dynamic model of aerial flight modes, it can be seen that the essence of the attitude change of the UAV is to eliminate the horizontal position error, that is:

[0124] (1);

[0125] In the formula, This refers to the horizontal position error, including errors in the x-axis and y-axis directions; This refers to the actual horizontal position. The target is located in a horizontal position.

[0126] For a horizontal position controller using PID control, the control law satisfies:

[0127] (2);

[0128] In the formula, , , These are the gain parameters for the proportional, integral, and derivative components of the horizontal position PID controller, respectively.

[0129] To achieve equation (3), the horizontal acceleration The following conditions must be met:

[0130] (3);

[0131] Therefore, we can conclude that:

[0132] (4);

[0133] In the formula: This is the attitude angle command; This is the angle transformation matrix.

[0134] The input to the inner-loop MPC attitude controller is the given target roll angle. Target pitch angle and target yaw angle The target attitude command is compared with the output of the prediction model collected by the feedback loop. The optimal control law is then used for online rolling calculation, i.e., solving the preset cost function, to obtain the optimal control sequence in the control time domain. The optimal control quantity in the control sequence at the current moment is then used as the output signal of the attitude controller. .

[0135] Based on the established nonlinear dynamics model of the in-flight flight of the reusable amphibious rotorcraft, a state-space equation is used to describe it, and its linear discrete state-space model is derived as follows:

[0136] (5);

[0137] In the formula, For the pose and state variables of the amphibious unmanned aerial vehicle; For output variables; To control the input amount; External disturbance quantity; A is the state matrix; B1 is the input control matrix; B2 is the disturbance control matrix; C is the output matrix; ignoring external input disturbance, set... The matrix parameters in equation (6) can be determined based on the dynamic parameters of the amphibious rotary-wing UAV.

[0138] By rewriting the linear discrete state-space model in incremental form to reduce the impact of static errors on the system's responsiveness, we have:

[0139] (6);

[0140] In the formula, These are the differences between state variables; To control the difference in scores;

[0141] To simplify the control code and adapt it to the hardware computing power and storage space of the underlying flight controller, the attitude controller in the amphibious UAV's airborne flight controller is designed as an MPC controller. Therefore, the state variables of the prediction model are selected as the attitude angles and angular velocities of the amphibious UAV, resulting in:

[0142] (7);

[0143] To predict the future flight attitude behavior of an amphibious UAV, a reasonable prediction equation needs to be constructed to optimize the future control trajectory. Treating the future system state variables and future control variables as adjustable variables, within a pre-selected prediction time domain, with k representing the current sampling time, the future control trajectory and state variables are described as follows:

[0144] (8);

[0145] In the formula, To control the step size; To predict the step size;

[0146] The predicted output can be represented as:

[0147] (9);

[0148] Based on equation (9), the future predicted output is represented in vector form as follows:

[0149] (10);

[0150] Separately , Then the above formula simplifies to:

[0151] (11);

[0152] By defining the boundaries between control inputs and state variables, a dynamic balance between control performance and physical constraints is achieved. When designing the MPC attitude controller, only the control variables are considered. and its rate of change Constraints. The constraint on the rate of change of the input is assigned to a single variable. During the sampling time, the following vector relationship is satisfied:

[0153] (12);

[0154] In the formula, To constrain the lower limit of the control quantity; The upper limit of the control quantity constraint conditions; the above two limiting parameters need to be determined in conjunction with the selection of motor and propeller;

[0155] The process of optimizing the flight attitude of an amphibious UAV using an MPC-PID dual closed-loop cascade controller can essentially be reduced to solving an optimization problem of a preset performance function. Therefore, a corresponding cost function needs to be defined to quantify the attitude control objective and solve for it. The optimal cost function is constructed by combining the weighted error between the predictive control output and the reference quantity with the weighted control quantity; solving for the optimal value of this function yields the optimal control sequence in the control time domain. For an amphibious UAV, its attitude control objective focuses on fast, accurate, and stable tracking of commands; therefore, the cost function of the MPC attitude controller is expressed as:

[0156] (13);

[0157] In the formula, The output weighting factor for the prediction error; Weighting factors to control the increment; The reference input sequence;

[0158] Therefore, the problem of controlling the flight attitude of amphibious UAVs can be reduced to an optimization problem of solving the following equation:

[0159] (14);

[0160] 2. Ground motion controller based on dual closed-loop PID

[0161] To achieve motion control of an amphibious rotary-wing UAV in ground mode, this invention employs dual closed-loop feedback control. Closed-loop feedback control is applied to the velocity and angular velocity of the power-reusable amphibious UAV during ground movement. Furthermore, PID controllers are added to both control loops to compensate for insufficient motion stability of the amphibious UAV under complex road surface conditions. The ground motion controller architecture based on dual closed-loop PID is shown in the attached figure. Figure 9 As shown.

[0162] The ground motion controller based on dual-loop PID uses the speed and yaw rate of the amphibious UAV as controlled variables. In the control closed loop, a PID controller is used to improve the error between the platform's real-time motion state and the target command. After conversion based on the ground wheel and rotor speeds, the actuators are driven to apply forces and torques to the fuselage, thereby achieving forward and turning control of the powered amphibious UAV in ground mode. The control law of the dual-loop PID ground motion controller satisfies:

[0163] (15);

[0164] In the formula, The speed control quantity output by the speed PID controller; , , These are the gain coefficients of the proportional, integral, and derivative components of the speed PID controller, respectively. The velocity error is in the ground mode. The angular velocity control quantity output by the yaw angular velocity PID controller; , , These are the gain coefficients of the proportional, integral, and derivative components of the yaw rate PID controller, respectively. This represents the yaw rate error in the ground mode.

[0165] The dual-loop PID controller, through the coordination of the speed loop and the angular velocity loop, can calculate the expected rotational speed of the ground wheels under the current control command. However, for the power-reusable amphibious rotorcraft UAV designed in this invention, its ground drive relies on the forward thrust generated by the tilted rotor, therefore, the expected rotational speed command of the ground wheels cannot be directly mapped to the rotor motor. A "ground wheel-rotor" rotational speed conversion stage needs to be added to convert the rotational speed command into an effective control signal suitable for driving the rotor motor.

[0166] Control quantities of velocity loop and angular velocity loop and Further processing yields the expected rotational speeds of the ground wheels on both sides, which satisfy the following conditions:

[0167] (16);

[0168] In the formula, The expected rotational speed of the left ground wheel; denoted as , where is the expected rotational speed of the right-side ground wheel; B is the diameter of the ground wheel; and r is the fuselage semi-wheelbase.

[0169] Assuming that the rotor motor drive signal and the expected rotational speed of the ground wheel satisfy a linear relationship, the drive control signals of the left and right rotor motors can be expressed as follows:

[0170] (17);

[0171] In the formula, This is the motor control signal for the left ground wheel; This is the motor control signal for the right-side ground wheel; This is the speed conversion factor.

[0172] After the drive control signals from the two ground wheels are transmitted to the motor controller, the actuator motor responds to the signals, driving the motor to rotate and generate forward thrust. Based on the response characteristics of a brushless DC motor, we can conclude that:

[0173] (18);

[0174] In the formula, , These represent the motor speeds in the left and right ground wheels, respectively. The time constant of the motor; This refers to the slope parameter of the motor curve; These are the parameters for the motor curve constants.

[0175] The relationship between the thrust and slip torque generated by the rotor motor and the motor speed can be expressed as:

[0176] (19);

[0177] In the formula, The thrust generated by the left ground wheel; The thrust generated by the right ground wheel; This is the ground wheel thrust coefficient; The slip moment generated by the ground wheel; The difference in thrust generated by the differential speed of the two ground wheels; This refers to the rotor tilt angle.

[0178] 3. Land-to-air mode switching controller integrating altitude information

[0179] This invention focuses on the characteristics of the land-air mode switching process and constructs a land-air switching control method that integrates altitude information to perform mode switching control of a power-reusable land-air amphibious UAV.

[0180] The process for achieving "land-to-air" switching is as follows: After receiving the mode switching command, the system first triggers the servo motor to drive the rotor tilt mechanism, adjusting the rotor tilt angle to complete the mode transition; simultaneously, the control system switches the underlying controller, automatically disabling the ground motion controller and switching to the airborne flight controller. After completing the mode switch, relying on the quadcopter's vertical takeoff and landing capabilities, it climbs to a certain altitude and flies over obstacles. Figure 10 As shown.

[0181] The process of switching from air to ground is as follows: The control system obtains real-time flight altitude through sensor modules such as barometers and GPS. When the flight altitude reaches a suitable threshold, the rotor tilts via servo motors, and the airborne flight controller is switched to the ground motion controller. Afterward, ground motion is achieved by driving the driven wheels with the rotor. Figure 11 As shown.

[0182] The land-to-air mode switching controller is the core component for enabling the interconnection and switching between the airborne flight controller and the ground motion controller of an amphibious UAV. It requires a logical switching mechanism to transmit control signals under specific modes to the actuators, while simultaneously switching the underlying motion controller. The basic architecture of the land-to-air switching controller designed in this invention is as follows: Figure 12As shown.

[0183] The control equations for the land-to-air handover controller can be expressed as follows:

[0184] (20);

[0185] In the formula, The control quantity that actually acts on the controlled object; The control quantity output by the flight motion controller; The control quantity output by the ground motion controller; This serves as a modal identifier, encompassing both air and ground modes. Furthermore, to achieve rotor tilt, an additional servo control channel is required. This is achieved through control variables. The servo motor is controlled by applying rotational torque after receiving a command. To achieve rotor tilting.

Claims

1. A control system for a reusable amphibious unmanned aerial vehicle (UAV), comprising: A remote control module used to apply control commands to the underlying control module, plan paths, monitor motion status in real time, and record status data; A communication module for sending control signals from the remote control module to the underlying control module and feeding back status information measured by the underlying control module to the remote control module; a drive module for driving the rotor motor to rotate and controlling its speed via an electronic speed controller or driving the rotor to tilt via a servo motor based on the control signals input from the underlying control module; characterized in that: the underlying control module includes a motion controller, a pressure gauge, an IMU, and a GPS, the motion controller having an airborne flight controller, a ground motion controller, and a land-air mode switching controller, wherein: The airborne flight controller includes a dual-closed-loop cascaded outer-loop PID position controller and an inner-loop MPC attitude controller, a control distributor, and actuators. The MPC attitude controller compares the target attitude command provided by the PID position controller with the output of the prediction model collected by the feedback correction loop, and performs online rolling calculations using the optimal control law, i.e., solving the preset cost function to obtain the optimal control sequence in the control time domain. The optimal control quantity in the control sequence at the current moment is used as the output signal of the MPC attitude controller. The output signals of the PID position controller and the MPC attitude controller are respectively input to the control distributor, and after distribution, the actuators are driven to perform actions to achieve airborne flight motion control. The ground motion controller includes a dual-loop PID controller, ground wheel and rotor speed conversion and actuator. The speed and yaw rate are used as inputs to the dual-loop PID controller. The dual-loop PID controller is used to improve the error between the real-time motion state and the target command. After the ground wheel and rotor speed is converted, the actuator is driven to perform actions, apply force and torque to the fuselage to achieve forward and turning control. The land-to-air mode switching controller includes a mode switching judgment and a servo controller. When the mode switching judgment receives a "land-to-air" mode switching command, it adjusts the rotor tilt angle through the servo controller to complete the land-to-air mode conversion and switches the ground motion controller to the air flight controller. When the real-time flight altitude obtained by the barometer, GPS and IMU sensors reaches the threshold, it adjusts the rotor tilt angle through the servo controller to complete the "air-to-land" mode conversion and switches the air flight controller to the ground motion controller.

2. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 1, characterized in that: The target attitude command is obtained in the following way: Based on the constructed nonlinear dynamics model of the amphibious rotorcraft UAV in flight: （1）； In the formula: It is a position vector; The velocity vector in the body coordinate system; The velocity vector in the inertial coordinate system; The coordinate rotation matrix; This is the vector of the body's rotational angular velocity; To combine external forces; This is the attitude angle vector; This is the attitude transformation matrix; It is the inertial tensor; The resultant torque; It can be seen that the essence of the attitude change of the UAV is to eliminate the horizontal position error, that is: (2)； In the formula: This refers to the horizontal position error, including errors in the x-axis and y-axis directions; This refers to the actual horizontal position. The target horizontal position; For a horizontal position controller using PID control, the control law satisfies: (3)； In the formula: , , These are the gain parameters for the proportional, integral, and derivative components of the horizontal position PID controller, respectively. To achieve equation (3), the horizontal acceleration The following conditions must be met: (4)； Therefore, we can conclude that: (5)； In the formula: This is the attitude angle command; This is the angle transformation matrix; Receive target attitude command This provides input commands to the MPC attitude controller in the attitude inner loop.

3. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 1, characterized in that: The working principle of the dual-closed-loop cascaded MPC-PID controller can be analyzed as a time-series recursive optimization process: at the current sampling time k, based on the embedded prediction equation and prediction model, the historical flight dataset and the executed pose control sequence are fused to construct k+1 to k+N. p The system state trajectory prediction at time N is obtained by optimizing the extreme value of the cost function and considering the boundary conditions of the motor's amplitude limit and state space constraints. C The optimal pose control solution sequence of step control is obtained, and the first quantity of the optimal solution sequence is injected into the control system as the execution quantity. When the system progresses to time k+1, the state observations are re-sampled for state estimation, thereby refreshing the predicted trajectory and solving the UAV pose optimization problem in the finite time domain again, forming a rolling optimization closed-loop architecture.

4. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 3, characterized in that: The specific process of the time-series recursive optimization process is as follows: Based on the established nonlinear dynamics model of the amphibious rotorcraft UAV in flight, a state-space equation is used to describe its linear discrete state-space model, which is then derived as follows: (6)； In the formula: For the pose and state variables of the amphibious unmanned aerial vehicle; For output variables; To control the input amount; External disturbance quantity; A is the state matrix; B1 is the input control matrix; B2 is the disturbance control matrix; C is the output matrix; Ignore external input interference and set The matrix parameters in equation (6) can be determined based on the dynamic parameters of the amphibious rotary-wing UAV. By rewriting the linear discrete state-space model in incremental form to reduce the impact of static errors on the system's responsiveness, we have: (7)； In the formula: These are the differences between state variables; To control the difference in scores; To simplify the complexity of the control code and adapt it to the hardware computing power and storage space of the underlying flight controller, the attitude controller in the amphibious UAV's airborne flight controller is designed as an MPC controller; therefore, the state variables of the prediction model are selected as the attitude angles and angular velocities of the amphibious UAV, which leads to: (8)； To predict the future flight attitude behavior of amphibious UAVs, a reasonable prediction equation needs to be constructed to optimize the future control trajectory. The future system state variables and future control variables are treated as adjustable variables. Within a pre-selected prediction time domain, where k represents the current sampling time, the future control trajectory and state variables are described as follows: (9)； In the formula: To control the step size; To predict the step size; The output of the pre-prediction is represented as: (10) Based on equation (9), the future predicted output is represented in vector form as follows: (11)； Separately , Then the above formula simplifies to: (12)； By defining the boundaries between control inputs and state variables, a dynamic balance between control performance and physical constraints is achieved. When designing the MPC attitude controller, only the control variables are considered. and its rate of change Constraints; assigning the constraint on the rate of change of input to a single variable. During the sampling time, the following vector relationship is satisfied: (13)； In the formula: To constrain the lower limit of the control quantity; This is the upper limit of the control quantity constraint condition; and It needs to be determined in conjunction with the selection of the motor and propeller; The process of optimizing the flight attitude of an amphibious UAV using a dual-closed-loop cascaded MPC-PID controller can essentially be reduced to solving an optimization problem of a preset performance function. Therefore, it is necessary to define a corresponding cost function to quantify the attitude control objective and solve it. The weighted error between the predictive control output and the reference quantity is combined with the weighted control quantity to construct the optimal cost function. Solving for the optimal value of this function yields the optimal control sequence in the control time domain. For amphibious unmanned aerial vehicles (UAVs), the attitude control objective focuses on fast, accurate, and stable tracking of commands. Therefore, the cost function of the MPC attitude controller is expressed as: (14)； In the formula: The output weighting factor for the prediction error; To control the weighting factor of the increment, The reference input sequence; Therefore, the problem of controlling the flight attitude of amphibious UAVs can be reduced to an optimization problem of solving the following equation: (15)； The optimal control sequence obtained by solving the above problem is used as the output signal of the MPC controller.

5. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 1, characterized in that: The dual closed-loop PID ground motion controller consists of a velocity loop PID controller that addresses the problem of lag in the motion speed response of amphibious UAVs in ground mode and an angular velocity loop PID controller that improves the problem of lag in the angular velocity response of amphibious UAVs in ground mode.

6. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 5, characterized in that: The control law of the dual closed-loop PID ground motion controller satisfies: (16)； In the formula: The speed control quantity output by the speed PID controller; , , These are the gain coefficients of the proportional, integral, and derivative components of the speed PID controller, respectively. The velocity error is in the ground mode. The angular velocity control quantity output by the yaw angular velocity PID controller; , , These are the gain coefficients of the proportional, integral, and derivative components of the yaw rate PID controller; This represents the yaw rate error in the ground mode.

7. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 6, characterized in that: The conversion of ground wheel and rotor speeds into the angular velocity control quantity output by the yaw rate PID controller is used. And the speed control quantity output by the speed PID controller This is converted into control signals to drive the rotor motor.

8. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 7, characterized in that: The specific process of obtaining the control signal involves: combining the control quantities of the velocity loop and the angular velocity loop... and Further processing revealed that the expected rotational speeds of the ground wheels on both the left and right sides met the following requirements: (17)； In the formula: The expected rotational speed of the left ground wheel; The expected rotational speed of the right ground wheel; B is the diameter of the ground wheel; r is the fuselage semi-wheelbase. Assuming that the rotor motor drive signal and the expected rotational speed of the ground wheel satisfy a linear relationship, the drive control signals of the left and right rotor motors can be expressed as follows: (18)； In the formula: This is the motor control signal for the left ground wheel; This is the motor control signal for the right-side ground wheel; This is the rotational speed conversion factor; After the drive control signals from the two ground wheels are transmitted to the motor controller, the actuator motor responds to the signals, driving the motor to rotate and generate forward thrust; based on the response characteristics of the brushless DC motor: (19)； In the formula: , These represent the motor speeds in the left and right ground wheels, respectively. The time constant of the motor; This refers to the slope parameter of the motor curve; These are the parameters for the motor curve constants; Based on the dynamic characteristics of the motor and the ground friction coefficient, the relationship between the thrust and slip torque generated by the rotor motor and the motor speed is expressed as follows: (20)； In the formula: The thrust generated by the left ground wheel; The thrust generated by the right ground wheel; This is the ground wheel thrust coefficient; The slip moment generated by the ground wheel; The difference in thrust generated by the differential speed of the two ground wheels; This refers to the rotor tilt angle.

9. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 1, characterized in that: The control equations for the land-air mode switching controller are expressed as follows: (21)； In the formula: The control quantity that actually acts on the controlled object; The control quantity output by the flight motion controller; The control quantity output by the ground motion controller; It is a modal identifier, including two states: air mode and land mode.

10. The control system for a power-reusable amphibious unmanned aerial vehicle according to claim 9, characterized in that: The servo controller controls the servo motor by using the control quantity provided by the control channel based on the mode switching judgment.

Images