Control method for ultra-low-altitude flight of large unmanned aerial vehicle

Abstract

The invention discloses a control method for ultra-low-altitude flight of a large unmanned aerial vehicle, and the method comprises the following steps: building a kinetic model of the unmanned aerial vehicle, the kinetic model comprising dynamic equations of position, speed, attitude control and control input; performing error tracking control on the attitude and the speed of the unmanned aerial vehicle by adopting a Backstepping nonlinear control method based on the dynamic model; estimating external disturbance and dynamic parameters of the unmanned aerial vehicle in real time by combining an L1 adaptive control method, and adaptively adjusting control input to compensate wind speed change; predicting a future flight state by using a model prediction control algorithm, optimizing a flight path and compensating wind speed disturbance; a safe trajectory is planned in real time through a dynamic window method, and obstacles are avoided in combination with dynamic constraints. According to the invention, through organic combination of Backstepping control, L1 adaptive control, MPC and DWA algorithms, the complex problems of wind speed disturbance, obstacle avoidance, high-precision operation and the like in ultra-low-altitude flight can be solved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) control, and particularly to a control method for the ultra-low altitude flight of large UAVs. Background Art

[0002] With the development of UAV technology, its application fields have been continuously expanded. In the fields of agriculture, forestry, etc., the demand for ultra-low altitude operations of large UAVs is increasing day by day. For example, in scenarios such as farmland or forest land, UAVs need to fly at ultra-low altitude when performing tasks. For instance, when carrying out pesticide spraying operations, to ensure the efficacy of pesticides and the accuracy of operations, the UAV needs to fly at a low altitude of 5 meters to achieve precise pesticide application to crops or trees. However, there are many obstacles such as tree belts and wires in such scenarios, which pose a serious threat to the flight safety of UAVs.

[0003] In the low-altitude environment, the wind speed is variable and complex. Strong winds will interfere with the flight attitude and trajectory of UAVs, affect their flight stability, and increase the control difficulty. Traditional flight control algorithms are difficult to cope with such complex and variable wind conditions and cannot meet the requirements for the stable flight of UAVs at ultra-low altitude. Therefore, a control method for the ultra-low altitude flight of large UAVs is needed to intelligently control UAVs.

[0004] Existing control methods cannot well control the low-altitude flight of UAVs. The influence of wind factors on the control effect is relatively large, which has a certain impact on the use of control methods. Therefore, a control method for the ultra-low altitude flight of large UAVs is proposed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: how to solve the problem that existing control methods cannot well control the low-altitude flight of UAVs, the influence of wind factors on the control effect is relatively large, and it has a certain impact on the use of control methods, and provide a control method for the ultra-low altitude flight of large UAVs.

[0006] The present invention solves the above technical problem through the following technical solutions. The present invention includes the following steps:

[0007] Establish a dynamic model of the UAV. The dynamic model includes dynamic equations of position, velocity, attitude control, and control input, and its specific content is as follows:

[0008]

[0009] Among them, the three-dimensional position vector of the UAV is expressed as r = [x, y, z] T , where x, y, and z are the positions of the UAV on the three spatial coordinate axes respectively, represents the rate of change of position;

[0010] The three-dimensional velocity vector of the UAV is expressed as v = [vx , v y , v z T , where v x , v y , v z are the velocities of the UAV on the three spatial coordinate axes respectively;

[0011] represents the rate of change of position; m is the mass of the UAV, F is the total thrust, and g is the gravitational acceleration vector;

[0012] Based on the dynamic model, the Backstepping nonlinear control method is used to perform error tracking control on the attitude and velocity of the UAV;

[0013] Combined with the L1 adaptive control method, the external disturbances and dynamic parameters of the UAV are estimated in real time, and the control input is adaptively adjusted to compensate for the wind speed change;

[0014] The model predictive control algorithm is used to predict the future flight state, optimize the flight path and compensate for the wind speed disturbance;

[0015] The safe trajectory is planned in real time through the dynamic window method, and obstacles are avoided by combining dynamic constraints.

[0016] Furthermore, the attitude control of the UAV is completed by adjusting the roll angle, pitch angle and yaw angle of the UAV;

[0017] The attitude error of the UAV is described by Euler angles, and the angular velocity and control torque of the UAV are used during the control process:

[0018]

[0019] where Θ = [φ, θ, ψ] T are the Euler angles of the UAV, φ, θ, ψ represent the roll angle, pitch angle and yaw angle of the UAV respectively, and ω = [ω x , ω y , ω z T is the angular velocity of the UAV;

[0020] The control torque of the UAV is related to the control input of the control surface. The control torque is related to the inertia matrix I of the UAV. The attitude control equation of the UAV is:

[0021]

[0022] where M is the control torque of the UAV and I is the inertia matrix of the UAV.

[0023] Furthermore, the Backstepping nonlinear control method includes:​​

[0024] Define the attitude error. The target attitude of the UAV is: Θ d =[φ d , θ d , θ, ψ] T , and the actual attitude is Θ. The attitude error e Θ =Θ d -Θ;

[0025] Construct the Lyapunov function, where P is a positive definite matrix representing the energy function of the attitude error;

[0026] Take the derivative of the Lyapunov function to obtain:

[0027] where, ω d and ω represent the target angular velocity and the actual angular velocity of the UAV respectively;

[0028] And through the control torque M θ =K1e Θ +K2ω, to perform stable attitude, where, K1 and K2 are gain matrices;

[0029] Define the velocity error, e v =v d -v, where, v d and v represent the target velocity and the actual velocity of the UAV respectively;

[0030] Through constructing the Lyapunov function for control, construct the Lyapunov function

[0031] where P2 is a positive definite matrix representing the energy function of the velocity error;

[0032] Take the derivative of the Lyapunov function to obtain:

[0033]

[0034] where, a d and a represent the target acceleration and the actual acceleration of the UAV respectively. Through the thrust control input u v =K3e v +K4a, to perform velocity tracking;

[0035] where, K3 and K4 are gain matrices.

[0036] Furthermore, the specific content of the L1 adaptive control method is as follows: Based on the state error e = [e θ , e v design the control law Update the dynamic parameters of the UAV in real time;

[0037] Among them, K1 and K2 are gain matrices, and e(t) represents the state error of the UAV, that is, the attitude error and the velocity error;

[0038] Based on the real-time estimation results, adjust the control input, evaluate the dynamics of the UAV in real time, and automatically adjust the control input according to environmental changes.

[0039] Furthermore, the content of the model predictive control algorithm is as follows:

[0040] Obtain wind speed information through sensors or estimation models, and introduce a wind speed disturbance compensation term in the optimization problem;

[0041] Through the formula Obtain the optimal control input solution and generate an anti-wind optimized trajectory;

[0042] Among them, u(t) represents the control input, including the thrust and moment control quantities;

[0043] e(t) represents the error between the target state and the actual state, and λ is a penalty factor used to balance the weights between the state error and the control input;

[0044] Solve the optimal control input by minimizing the cost function, minimize the control input, reduce the state error at the same time, and compensate for the wind speed disturbance through the optimal control problem;

[0045] The model predictive control algorithm considers the dynamic model of the UAV, the obstacle positions, and the control input constraints during the optimization process, generates the optimal path, and dynamically adjusts the flight trajectory.

[0046] Furthermore, the specific content of the dynamic window method is as follows:

[0047] An algorithm for optimizing the flight path based on real-time environmental information, which determines the safe trajectory selected by the UAV by calculating the dynamic constraints of the UAV using the dynamic window method;

[0048] The control input range of the UAV is determined by the constraints of the current speed and acceleration. Select a dynamic window, and then optimize and select the flight path based on the trajectories in the selected dynamic window;

[0049] The UAV calculates the path in real time through the dynamic window method and avoids obstacles during flight;

[0050] The goal is to ensure that the UAV can safely avoid obstacles such as tree belts and power lines while ensuring flight efficiency.

[0051] The present invention has the following advantages compared with the prior art: For the control method of the large unmanned aerial vehicle (UAV) flying at ultra-low altitude, the UAV combines Backstepping control and L1 adaptive control, which has advantages in stability and robustness.

[0052] Combined with DWA: In terms of flight path optimization and obstacle avoidance, a strategy combining Model Predictive Control (MPC) and Dynamic Window Approach (DWA) is adopted, which can adjust the path in real time, avoid obstacles, and cope with environmental changes.

[0053] Wind speed disturbance compensation during ultra-low altitude flight: Through the optimization of L1 adaptive control and MPC algorithms, the UAV realizes the compensation for wind speed changes, ensuring stability.

[0054] The UAV can perform ultra-low altitude spraying tasks in complex environments, with extremely high stability and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is the overall flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] The following provides a detailed description of the embodiments of the present invention. These embodiments are implemented on the premise of the technical solution of the present invention, and detailed implementation manners and specific operation processes are given. However, the protection scope of the present invention is not limited to the following embodiments.

[0057] As Figure 1 shown, this embodiment provides a technical solution: a control method for a large UAV flying at ultra-low altitude, including the following steps:

[0058] Before designing the control algorithm, it is first necessary to model the dynamics of the UAV;

[0059] Construct the motion equation of the UAV:

[0060] The dynamic system of the UAV can be described by Newton-Euler equations, and the state of the UAV includes position, velocity, and attitude.

[0061] The dynamic model of the UAV can be written as follows:

[0062]

[0063]

[0064] where r = [x, y, z] T is the three-dimensional position of the UAV, v = [v x , v y , v z T$v$ is the speed of the UAV, $m$ is the mass of the UAV, $F$ is the total thrust, and $g$ is the gravitational acceleration vector;

[0065] $\frac{d}{dt}$ represents the derivative of a vector, which is a mathematical formula, and $T$ represents the transpose of a vector matrix, both of which are commonly used symbolic representations in mathematics.

[0066] Construction of attitude and angular velocity equations:

[0067] The attitude control of the UAV is achieved by adjusting the roll angle, pitch angle, and yaw angle of the UAV. The attitude error of the UAV can be described by Euler angles. The angular velocity and control torque of the UAV are used in the control process:

[0068]

[0069] where $\theta = [\varphi,\theta,\psi]$ T is the Euler angle (roll angle, pitch angle, yaw angle) of the UAV, $\omega = [\omega$ x , $\omega$ y , $\omega$ z $ T is the angular velocity of the UAV, $\dot{\theta}$ represents the rate of change of the Euler angle of the UAV.

[0070] The control torque of the UAV is related to the control input of the control surface, and these torques are related to the inertia matrix $I$ of the UAV. The attitude control equation of the UAV is:

[0071]

[0072] where $M$ is the control torque of the UAV and $I$ is the inertia matrix of the UAV.

[0073] Relationship between control input and thrust:

[0074] The control input of the UAV includes thrust and torque. The design goal of the control input is to minimize the error between the actual motion state and the desired state of the UAV. In ultra-low altitude flight, the calculation of the thrust $T$ and torque $M$ is based on the angle of the control surface and the aerodynamic model.

[0075] Flight control algorithm design:

[0076] The core task of the flight control algorithm is to maintain the stability of the UAV in a complex environment and complete the task under the influence of variable wind speeds, flight altitudes, and obstacles. To achieve this goal, this algorithm design will adopt a control strategy that combines Backstepping control, L1 adaptive control, model predictive control (MPC), and dynamic window approach (DWA).

[0077] Backstepping nonlinear control:

[0078] Backstepping is a method for gradually constructing a controller, which is suitable for dealing with nonlinear systems. In flight control, Backstepping is usually applied to attitude control and speed control. Its basic idea is to ensure the stability of the system by constructing Lyapunov functions and in a recursive manner. Combining with the dynamic model of the UAV, the control designs for attitude and speed are as follows:

[0079] Attitude control:

[0080] The target attitude of the UAV is: θ d , the actual attitude is θ, and the attitude error e θ = θ d - θ.

[0081] Define the Lyapunov function as:

[0082]

[0083] where P is a positive definite matrix.

[0084] Take the derivative of the Lyapunov function to get:

[0085]

[0086] To ensure the stability of the system, the control input M θ is designed as:

[0087] M θ = K1e θ + K2ω

[0088] where K1 and K2 are gain matrices.

[0089] Speed control:

[0090] The speed error e v = v d - v can be controlled by constructing a Lyapunov function:

[0091]

[0092] Take the derivative of the Lyapunov function to get:

[0093]

[0094] The control input u v is:

[0095] u v = K3e v + K4a

[0096] This control input enables the UAV to track the target speed.

[0097] L1 Adaptive Control:

[0098] L1 adaptive control is a control method used to handle system uncertainties and external disturbances. Its core idea is to estimate the state and dynamic parameters of the UAV in real time and adaptively adjust the control input.

[0099] Error Model and Control Law:

[0100] The state error e of the UAV = [e θ , e v , which is corrected through the L1 control law:

[0101] where K1 and K2 are gain matrices, is the rate of change of the error. The L1 control algorithm can estimate the dynamics of the UAV in real time and automatically adjust the control input according to environmental changes, especially suitable for compensating for wind speed changes and external disturbances during flight.

[0102] Adaptive Regulation and State Estimation:

[0103] L1 adaptive control can update the dynamic parameters of the UAV in real time and adjust the control input based on the real-time estimation results to ensure the UAV remains stable under disturbances and uncertainties.

[0104] Model Predictive Control (MPC):

[0105] Model Predictive Control (MPC) is a forward-looking control method that uses the dynamic model of the UAV to predict future states and control inputs. MPC adjusts the trajectory of the UAV by solving an optimal control problem and takes into account the constraints of the system.

[0106] Wind Speed Estimation and Compensation:

[0107] The MPC algorithm first obtains wind speed disturbance information through sensors or estimation models and considers the influence of wind speed during the optimization process. The control objective is:

[0108]

[0109] where e(t) is the state error of the UAV, λ is the penalty factor, and the control input compensates for wind speed disturbances through an optimal control problem.

[0110] Optimizing Flight Path and Obstacle Avoidance:

[0111] During the optimization process, the MPC algorithm takes into account the dynamic model of the UAV, the positions of obstacles, and control input constraints, generates an optimal path, and dynamically adjusts the flight trajectory to ensure that the UAV avoids obstacles and copes with complex environments.

[0112] Dynamic Window Approach (DWA):

[0113] The Dynamic Window Approach (DWA) is an algorithm that optimizes the flight path based on real-time environmental information (such as obstacles, wind speed, etc.). By calculating the dynamic constraints of the UAV (speed, acceleration, etc.), DWA can determine the safe trajectories that the UAV can choose.

[0114] Dynamic Window Selection:

[0115] In DWA, the control input range of the UAV is determined by the constraints of the current speed and acceleration. A dynamic window is selected, and then the flight path is optimized and selected based on the trajectories within this window.

[0116] Path Planning and Obstacle Avoidance:

[0117] The UAV calculates the path in real time through DWA and avoids obstacles during flight. The goal is to ensure that the UAV can safely avoid obstacles such as tree belts and power lines while guaranteeing flight efficiency.

[0118] 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 quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0119] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0120] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. Control method for ultra-low altitude flight of large unmanned aerial vehicles, characterized in that, It includes the following steps: Establish a dynamic model of the UAV, which includes dynamic equations of position, velocity, attitude control, and control input. The specific content is as follows: Among them, the three-dimensional position vector of the drone is expressed as r = [x, y, z] T , where x, y, and z are the positions of the drone on the three spatial coordinate axes respectively, representing the rate of change of the position; The three-dimensional velocity vector of the drone is expressed as v = [v x , v y , v z T , where v x , v y , v z are the velocities of the drone on the three spatial coordinate axes respectively;​ represents the rate of change of position; m is the mass of the UAV, F is the total thrust, and g is the gravitational acceleration vector; Based on the dynamic model, use the Backstepping nonlinear control method to perform error tracking control on the attitude and velocity of the UAV; Combine the L1 adaptive control method to estimate the external disturbance and dynamic parameters of the UAV in real time, and adaptively adjust the control input to compensate for the wind speed change; Use the model predictive control algorithm to predict the future flight state, optimize the flight path, and compensate for the wind speed disturbance; Real-time plan a safe trajectory through the dynamic window method, and combine dynamic constraints to avoid obstacles.

2. The control method for the ultra-low altitude flight of a large unmanned aerial vehicle according to claim 1, characterized in that: Complete the attitude control of the UAV by adjusting the roll angle, pitch angle, and yaw angle of the UAV; Describe the attitude error of the UAV through Euler angles, and in the control process, use the angular velocity and control torque of the UAV: where Θ = [φ, θ, ψ] T is the Euler angle of the UAV, and φ, θ, ψ respectively represent the roll angle, pitch angle, and yaw angle of the UAV, ω = [ω x , ω y , ω z T is the angular velocity of the UAV;​ The control torque of the UAV is related to the control input of the rudder surface. The control torque is related to the inertia matrix I of the UAV. The attitude control equation of the UAV is: Where M is the control torque of the UAV, and I is the inertia matrix of the UAV.

3. The control method for ultra-low altitude flight of a large unmanned aerial vehicle according to claim 1, characterized in that: The Backstepping nonlinear control method includes: Define the attitude error. The target attitude of the UAV is: Θ d = [φ d , θ d , θ, ψ] T , the actual attitude is Θ, and the attitude error e Θ = Θ d - Θ; Construct the Lyapunov function, where P is a positive definite matrix, representing the energy function of the attitude error; Taking the derivative of the Lyapunov function, we get: where ω d and ω respectively represent the target angular velocity and the actual angular velocity of the UAV; and by controlling the moment M θ = K1e Θ + K2ω to perform attitude stabilization, where K1 and K2 are gain matrices; Define the speed error, e v = v d - v, where v d and v respectively represent the target speed and the actual speed of the UAV; Control is carried out by constructing a Lyapunov function, and the Lyapunov function is constructed Where P2 is a positive definite matrix, representing the energy function of the velocity error; Take the derivative of the Lyapunov function to obtain: where a d and a respectively represent the target acceleration and the actual acceleration of the UAV, and the thrust control input u v = K3e v + K4a is used for speed tracking; Where K3 and K4 are gain matrices.

4. The control method for ultra-low altitude flight of a large unmanned aerial vehicle according to claim 1, characterized in that: The specific content of the L1 adaptive control method is as follows: Based on the state error e = [e θ , e v , design the control law Update the dynamic parameters of the UAV in real time; Where K1 and K2 are gain matrices, and e(t) represents the state error of the UAV, that is, the attitude error and velocity error; Adjust the control input based on the real-time estimation result, evaluate the dynamics of the UAV in real time, and automatically adjust the control input according to the environmental change.

5. The control method for ultra-low altitude flight of a large unmanned aerial vehicle according to claim 1, characterized in that: The content of the model predictive control algorithm is as follows: Obtain the wind speed information through sensors or estimation models, and introduce a wind speed disturbance compensation term in the optimization problem; Obtained the solution of the optimal control input through the formula and generated the wind-resistant optimization trajectory; Where u(t) represents the control input, including the thrust and torque control quantities; e(t) represents the error between the target state and the actual state, and λ is a penalty factor used to balance the weight between the state error and the control input; Solve the optimal control input by minimizing the cost function, so that the control input is minimized, while reducing the state error, and compensate for the wind speed disturbance through the optimal control problem; The model predictive control algorithm considers the dynamic model of the UAV, the position of the obstacle, and the control input constraints during the optimization process, generates the optimal path, and dynamically adjusts the flight trajectory.

6. The control method for the ultra-low altitude flight of a large unmanned aerial vehicle according to claim 1, characterized in that: The specific content of the dynamic window method is as follows: An algorithm for optimizing the flight path based on real-time environmental information. By calculating the dynamic constraints of the UAV, use the dynamic window method to determine the safe trajectory selected by the UAV; The control input range of the UAV is determined by the constraints of the current speed and acceleration. Select a dynamic window, and then optimize and select the flight path based on the trajectory in the selected dynamic window; The UAV calculates the path in real time through the dynamic window method and avoids obstacles during flight.

Images