Flight control method for path tracking of coaxial dual-rotor unmanned aerial vehicle and unmanned aerial vehicle
By establishing a dynamic model and an adaptive robust trajectory tracking control method, combined with an RBFNN attitude and position controller, the problem of accurate attitude and position tracking for coaxial dual-rotor UAVs was solved, achieving precise and stable trajectory tracking control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINESE PEOPLES LIBERATION ARMY UNIT 32180
- Filing Date
- 2024-08-27
- Publication Date
- 2026-06-26
AI Technical Summary
Existing coaxial dual-rotor UAVs struggle to achieve precise attitude adjustment and position tracking, hampered by inaccurate dynamic models and external interference, which affects their flexible and accurate mission execution.
A dynamic model based on the Euler-Poincaré equations is established, and an adaptive robust trajectory tracking control method is adopted. Combined with an adaptive radial basis neural network (RBFNN) attitude and position controller, unknown disturbances are compensated to improve the robustness of the controller.
It achieves precise and stable trajectory tracking control of coaxial dual-rotor UAVs, improving anti-interference performance and control effect.
Smart Images
Figure CN122284633A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a flight control method for tracking the trajectory of a coaxial dual-rotor UAV and the UAV itself. Background Technology
[0002] Coaxial rotor aircraft (CRAs), as a special type of rotorcraft, possess outstanding characteristics such as small size, light weight, easy portability, and flexible flight, and have broad application prospects. Structurally, CRAs are compact, employing folding blades for easy carrying and storage. In terms of flight efficiency, the upper and lower rotors rotate in opposite directions, canceling out torque and easily achieving a good self-stabilizing state. Regarding takeoff and landing conditions, CRAs can take off and land vertically from the ground or other confined spaces, making them extremely convenient to use. However, limited by inaccurate dynamic models and external interference, current coaxial rotor aircraft still struggle to achieve precise attitude adjustment and position tracking, severely restricting their ability to flexibly and accurately achieve ideal mission results. Summary of the Invention
[0003] This application provides a flight control method and a UAV for track tracking of a coaxial dual-rotor UAV, which achieves precise and stable track tracking control of the CRA by improving the accuracy of the mathematical model and the robustness of the CRA controller.
[0004] This application provides a flight control method for tracking the trajectory of a coaxial dual-rotor unmanned aerial vehicle (UAV), including:
[0005] A dynamic model of the coaxial dual-rotor UAV CRA is established based on the Euler-Poincaré equations.
[0006] Based on the flight characteristics of CRA trajectory tracking, the control objective of the CRA adaptive robust trajectory tracking control method is established;
[0007] Based on the control objective, an attitude controller based on an adaptive radial basis function neural network (RBFNN) is configured; and,
[0008] Based on the motion characteristics of CRA trajectory tracking, a position controller based on adaptive parameter estimation is set up.
[0009] The coaxial dual-rotor UAV is controlled using the attitude controller and the position controller.
[0010] Optionally, based on the Euler-Poincaré equations, the dynamic model of the coaxial dual-rotor UAV CRA can be established as follows:
[0011] The dynamic model of CRA established based on the Euler-Poincaré equations satisfies:
[0012]
[0013] in, s() represents sin(), and c() represents cos();
[0014] The thrust of the drone satisfies:
[0015] F = [U2D U3D U4 U3 U2 U1] T +V·[0 0 0 0 0 -mg] T
[0016] Where U1 = T1 + T 2z U2 = T 2y U3 = T 2x U4 = ξ(T1-T2), where U1 represents the control parameter in the vertical direction; U2 represents the control parameter in the roll direction; U3 represents the control parameter in the pitch direction; U4 represents the control parameter in the yaw direction; x represents the control parameter in the rotation direction; D represents the length from the lower rotor of the UAV to the center of the UAV; and T... x T y T z This represents the components of rotor thrust in the three directions of the airframe coordinate system;
[0017] The thrust of the rotor system must satisfy:
[0018]
[0019] In the formula, α represents the aerodynamic angle of attack of the rotor, ρ0 represents the gas density in the flight airspace, and d l (x) represents the chord length of the blade section at a certain blade element, C T Ω represents the lift coefficient, and Ω represents the rotational speed of the UAV propeller blades.
[0020] Based on the CRA's dynamic model, and considering the model's uncertainties and external disturbances, the optimized dynamic model satisfies:
[0021]
[0022] Where u1=U1 / m, u2=U2 / Ix, u3=U3 / I y u4=U4 / I z d x ,d y ,d z ,d φ ,d θ ,d ψ This represents the unknown disturbance in each channel.
[0023] Optionally, based on the flight characteristics of CRA trajectory tracking, the control objectives of establishing the CRA adaptive robust trajectory tracking control method include:
[0024] The desired attitude angle of the CRA is Θ. d =[φ d θ d ψ d ] T The attitude angle error satisfies:
[0025]
[0026] The required roll and pitch angles are obtained by inverse kinematics of the position model:
[0027]
[0028] The rotational motion model then satisfies:
[0029]
[0030] Where, d Θ =[d x d y d z ] T u represents the unknown disturbance compensation amount. Θ =[u2 u3 u4] T This indicates the control input item.
[0031] Optionally, according to the control objective, setting an attitude controller based on an adaptive radial basis function neural network (RBFNN) includes:
[0032] For unknown disturbance quantities Provide compensation, take Use RBFNN to approximate the function f Θ =[f φ f θ f ψ ] T The uncertainty of the model is estimated and compensated in real time.
[0033]
[0034] Design an adaptive RBFNN network, defining the tracking error of the attitude system as follows:
[0035] e = Θ d -Θ
[0036] Define the auxiliary state error vector E as follows:
[0037]
[0038] Set an adaptive RBFNN to approximate the nonlinear unknown disturbance term f. Θ To obtain the output of the RBFNN network that satisfies:
[0039]
[0040] in, This represents the output vector of the RBFNN; This represents the input vector of the RBFNN, namely the variables such as the desired attitude angle and the actual value mentioned above. This is the network weight coefficient matrix; J > 1 represents the number of hidden layer nodes in the network. For vector mapping, each function H j It is a Gaussian function;
[0041]
[0042] Where i is the number of input nodes in the network, j represents the j-th node in the hidden layer of the network, and c ij σ represents the position of the node. j Let be the width of the j-th neuron. The optimal weight values set for the network;
[0043] The output of the RBFNN network is:
[0044] f Θ (Θ in ) = (W * ) T H(Θ in )+ε
[0045] Where ε is the neural network approximation error, satisfying ε max >0.
[0046] Optionally, according to the control objective, setting the attitude controller based on the adaptive radial basis function neural network (RBFNN) further includes:
[0047] Considering the horizontal motion model of CRA and the structure of RBFNN, design the control law:
[0048]
[0049] Design an adaptive algorithm to adjust the weights of the RBFNN network online, with the control law satisfying:
[0050]
[0051] Among them, K Θ It is a matrix of positive constants.
[0052] Optionally, based on the motion characteristics of CRA trajectory tracking, a position controller based on adaptive parameter estimation is configured, including:
[0053] Based on the drone's location and desired location Determine the position error:
[0054]
[0055] Define an auxiliary function:
[0056]
[0057] Among them, Γ1=diag[λ1 λ2 λ3];
[0058] Speed error is defined as:
[0059]
[0060] Set position control input:
[0061]
[0062] in, For k p The estimated value, For d p The estimated value is Γ2 = diag[λ4 λ5 λ6];
[0063] The update rules for the estimated displacement resistance coefficient and the unknown external disturbance value in the design control input satisfy the following:
[0064]
[0065]
[0066] in,
[0067] This application also proposes a coaxial dual-rotor unmanned aerial vehicle (UAV), including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the steps of the flight control method for tracking the trajectory of the coaxial dual-rotor UAV as described above.
[0068] This application proposes a control method that is simple to implement, has good control effect, and strong anti-interference performance. By improving the accuracy of the mathematical model and the robustness of the CRA controller, precise and stable trajectory tracking control of CRA is achieved.
[0069] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description
[0070] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:
[0071] Figure 1 This is a schematic diagram of the CRA model for an embodiment of this application;
[0072] Figure 2 This is a schematic diagram of the control flow of the CRA trajectory tracking control method in the embodiments of this application;
[0073] Figure 3 The CRA trajectory tracking control method in this application is based on the attitude control structure of RBFNN.
[0074] Figure 4 The flight trajectory in the numerical simulation of the CRA trajectory tracking control method in the embodiments of this application;
[0075] Figure 5 The position curve is shown in the numerical simulation of the CRA trajectory tracking control method in the embodiments of this application. Detailed Implementation
[0076] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0077] To achieve precise and stable trajectory tracking control for CRA (Current Radial Radial Function Neural Network), considering the advantages of Radial Basis Function Neural Network (RBFNN) in modeling, disturbance estimation, and compensation, this application provides a flight control method for trajectory tracking of a coaxial dual-rotor UAV, such as... Figure 1 As shown, it includes the following steps:
[0078] In step S101, a dynamic model of the coaxial dual-rotor UAV CRA is established based on the Euler-Poincaré equations. Figure 1This is a schematic diagram of the CRA model. Figure 1 The parameters defined in the code are as follows: V = [pqruvw] T Let be the velocity vector of the UAV in the body coordinate system; p, q, r are the angular velocities of the UAV in the three rotational directions in the body coordinate system; u, v, w are the velocities of the UAV in the body coordinate system. Position is... posture as Where f, θ, and ψ represent the roll angle, pitch angle, and yaw angle, respectively. T1 and T2 are the thrusts of the upper and lower rotors, respectively, m is the mass of the UAV, g is the acceleration due to gravity, and I... x I y I z It represents the axial rotational inertia.
[0079] In some embodiments, based on the aerodynamic characteristics and flight principles of the CRA, and using the Euler-Poincaré equations, a dynamic model of the coaxial dual-rotor unmanned aerial vehicle (UAV) CRA is established, including:
[0080] The dynamic model of CRA established based on the Euler-Poincaré equations satisfies:
[0081]
[0082] in, s() represents sin(), and c() represents cos();
[0083] The thrust of the drone satisfies:
[0084] F = (U2D U3D U4 U3 U2 U1) T +v·[0 0 0 0 0 -mg] T (2)
[0085] Where U1 = T1 + T 2z U2 = T 2y U3 = T 2x U4 = ξ(T1-T2), where U1 represents the control parameter in the vertical direction; U2 represents the control parameter in the roll direction; U3 represents the control parameter in the pitch direction; U4 represents the control parameter in the yaw direction; x represents the control parameter in the rotation direction; D represents the length from the lower rotor of the UAV to the center of the UAV; and T... x T y T z This represents the components of rotor thrust in the three directions of the airframe coordinate system;
[0086] The thrust of the rotor system must satisfy:
[0087]
[0088] In the formula, α represents the aerodynamic angle of attack of the rotor, ρ0 represents the gas density in the flight airspace, and d l (x) represents the chord length of the blade section at a certain blade element, C T Ω represents the lift coefficient, and Ω represents the rotational speed of the UAV propeller blades.
[0089] Based on the CRA's dynamic model, the optimized dynamic model satisfies:
[0090]
[0091] Considering the uncertainty of the model and external disturbances, corresponding to each channel of the control model, the dynamic equation (4) of CRA is described as follows:
[0092]
[0093] Where u1 = U1 / m, u2 = U2 / I x u3=U3 / I y u4=U4 / I z d x d y d z d φ d θ d ψ This represents the unknown perturbation (including model uncertainty and external perturbation) for each channel.
[0094] In step S102, based on the flight characteristics of CRA trajectory tracking, the control objective of the CRA adaptive robust trajectory tracking control method is established.
[0095] In step S103, an attitude controller based on an adaptive radial basis function neural network (RBFNN) is set according to the control objective.
[0096] In step S104, a position controller based on adaptive parameter estimation is set according to the motion characteristics of CRA trajectory tracking.
[0097] In step S105, the coaxial dual-rotor UAV is controlled using the attitude controller and the position controller.
[0098] This application proposes a control method that is simple to implement, has good control effect, and strong anti-interference performance. By improving the accuracy of the mathematical model and the robustness of the CRA controller, precise and stable trajectory tracking control of CRA is achieved.
[0099] In some embodiments, based on the flight characteristics of CRA trajectory tracking, the control objectives of establishing the CRA adaptive robust trajectory tracking control method include:
[0100] The desired attitude angle of the CRA is Θ. d =[φ d θ d ψ d ] T The attitude angle error satisfies:
[0101]
[0102] The required roll and pitch angles are obtained by inverse kinematics of the position model:
[0103]
[0104] Considering the rotational motion model of CRA and the unknown disturbance, the attitude model, i.e., the rotational motion model, in formula (7) can be expressed as:
[0105]
[0106] Where, d Q =[d x d y d z ] T u represents the unknown disturbance compensation amount. Q =[u2 u3 u4] T This indicates the control input item.
[0107] In some embodiments, setting an attitude controller based on an adaptive radial basis function neural network (RBFNN) according to the control objective includes:
[0108] For unknown disturbance quantities Provide compensation, take Use RBFNN to approximate the function f Θ =[f φ f θ f ψ ] T The uncertainty of the model is estimated and compensated in real time. The attitude model in equation (7) can be written in the form of a second-order nonlinear system:
[0109]
[0110] Design an adaptive RBFNN network, defining the tracking error of the attitude system as follows:
[0111] e = Θ d -Θ (10)
[0112] Define the auxiliary state error vector E as follows:
[0113]
[0114] Set an adaptive RBFNN to approximate the nonlinear unknown disturbance term f. Θ And to ensure that the tracking error eventually converges to a sufficiently small region, the neural network attitude control structure is as follows: Figure 2 As shown, the output of the RBFNN network satisfies:
[0115]
[0116] in, This represents the output vector of the RBFNN; This represents the input vector of the RBFNN, namely the variables such as the desired attitude angle and the actual value mentioned above. This is the network weight coefficient matrix; J > 1 represents the number of hidden layer nodes in the network. For vector mapping, each function H j It is a Gaussian function;
[0117]
[0118] Where i is the number of input nodes in the network, j represents the j-th node in the hidden layer of the network, and c ij σ represents the position of the node. j W is the width of the j-th neuron. i * The optimal weight values set for the network;
[0119] The output of the RBFNN network is:
[0120] f Θ (Θ in ) = (W * ) T H(Θ in )+ε (14)
[0121] Where ε is the neural network approximation error, satisfying ε max >0.
[0122] In some embodiments, setting an attitude controller based on an adaptive radial basis function neural network (RBFNN) according to the control objective further includes:
[0123] Considering the horizontal motion model of CRA and the structure of RBFNN, design the control law:
[0124]
[0125] Design an adaptive algorithm to adjust the weights of the RBFNN network online, with the control law satisfying:
[0126]
[0127] Among them, K Θ It is a matrix of positive constants.
[0128] In some embodiments, setting a position controller based on adaptive parameter estimation according to the motion characteristics of CRA trajectory tracking includes:
[0129] Based on the drone's location and desired location Determine the position error:
[0130]
[0131] Define an auxiliary function:
[0132]
[0133] Among them, Γ1=diag[λ1 λ2 λ3];
[0134] Speed error is defined as:
[0135]
[0136] Set position control input:
[0137]
[0138] in, For k p The estimated value, For d p The estimated value is Γ2 = diag[λ4 λ5 λ6];
[0139] The update rules for the estimated displacement resistance coefficient and the unknown external disturbance value in the design control input (20) satisfy formulas (21) and (22):
[0140]
[0141]
[0142] in,
[0143] The three-dimensional curve of the CRA's flight path is as follows: Figure 3 As shown, the specific position change curve is as follows: Figure 4As shown in the figures, combining the curves in the two graphs to compare the control effects of the two control methods reveals that the adaptive RBFNN control method achieves good overlap between the flight trajectory and the desired path, with fast convergence. Since the initial and desired position coordinates coincide in the x-direction, no difference in convergence speed between the two control methods is observed in the x-direction; however, the difference is more pronounced in the y-direction, with both methods reaching convergence in 1 second. After convergence, the deviation in each coordinate direction is small, enabling precise and stable flight tracking.
[0144] CRA posture changes such as Figure 5 As shown, under adaptive RBFNN control, the desired attitude angle can be tracked. The tracking errors for both roll and pitch angles are within 0.001 rad. In terms of tracking accuracy, especially near the peak of the attitude angle change, such as at 4 seconds, the roll angle under adaptive RBFNN control can still track the desired angle well. This phenomenon also exists in pitch and yaw angles, such as at 6 seconds and 18 seconds in the pitch angle change curve, and at 6 seconds and 12 seconds in the yaw angle change curve. The maximum error for pitch angle is 0.2 × 10⁻² rad, and the maximum error for yaw angle is 1 × 10⁻² rad. The initial values of roll and pitch angles are consistent with the initial values of the desired angles, indicating good convergence. However, the initial value of yaw angle is inconsistent with the desired value. From the tracking results, it can be seen that the yaw angle tracking performance after convergence shows no significant difference. Similar to the phenomena exhibited by position tracking, the RBFNN control method can achieve precise and stable attitude control, and can achieve good stability and attitude tracking performance. It can quickly control the attitude to converge to the desired attitude from the initial point.
[0145] Based on the aerodynamic characteristics and flight principles of the CRA, this application establishes a dynamic model of the CRA, considering the impact of unknown external disturbances on the UAV within the dynamic model. Then, position and attitude channel controllers are designed separately. For unknown external disturbances, an adaptive neural network is designed to estimate the disturbances and compensate the controllers, thereby improving the robustness and flight performance of the UAV.
[0146] This application also proposes a coaxial dual-rotor unmanned aerial vehicle (UAV), including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the steps of the flight control method for tracking the trajectory of the coaxial dual-rotor UAV as described above.
[0147] It should be noted that, in the embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.
[0148] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0149] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0150] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims. All of these forms are within the protection scope of this application.
Claims
1. A flight control method for tracking the trajectory of a coaxial dual-rotor unmanned aerial vehicle (UAV), characterized in that, include: A dynamic model of the coaxial dual-rotor UAV CRA is established based on the Euler-Poincaré equations. Based on the flight characteristics of CRA trajectory tracking, the control objective of the CRA adaptive robust trajectory tracking control method is established; Based on the control objective, an attitude controller based on an adaptive radial basis function neural network (RBFNN) is configured. as well as, Based on the motion characteristics of CRA trajectory tracking, a position controller based on adaptive parameter estimation is set up. The coaxial dual-rotor UAV is controlled using the attitude controller and the position controller.
2. The flight control method for tracking the trajectory of a coaxial dual-rotor UAV as described in claim 1, characterized in that, Based on the Euler-Poincaré equations, the dynamic model of the coaxial dual-rotor UAV CRA is established, including: The dynamic model of CRA established based on the Euler-Poincaré equations satisfies: in, s() represents sin(), and c() represents cos(); The thrust of the drone satisfies: F = [U2D U3D U4 U3 U2 U1] T + v - [0 0 0 0 0 -mg] T wherein U1 = T1 + T 2z , U2 = T 2y , U3 = T 2x , U4 = ξ(T1 - T2), U1 represents a control parameter in the vertical direction; U2 represents a control parameter in the roll direction; U3 represents a control parameter in the pitch direction; U4 represents a control parameter in the yaw direction; x represents a control parameter in the rotation direction; D represents the length from the lower rotor of the unmanned aerial vehicle to the center of the unmanned aerial vehicle, T x , T y , T z represent the components of the rotor thrust in the three directions of the body coordinate system; The thrust of the rotor system must satisfy: In the formula, a represents the aerodynamic angle of attack of the rotor, r0 represents the gas density in the flight airspace, and d l (x) represents the chord length of the blade section at a certain blade element, C T represents the lift coefficient, and w represents the rotational speed of the UAV propeller blades; Based on the CRA's dynamic model, and considering the model's uncertainties and external disturbances, the optimized dynamic model satisfies: Where u1 = U1 / m, u2 = U2 / I x u3=U3 / I y u4=U4 / I z d x d y d z d φ d θ d ψ This represents the unknown disturbance in each channel.
3. The flight control method for tracking the trajectory of a coaxial dual-rotor UAV as described in claim 2, characterized in that, Based on the flight characteristics of CRA trajectory tracking, the control objectives of establishing the CRA adaptive robust trajectory tracking control method include: The desired attitude angle of the CRA is Θ. d =[φ d θ d ψ d ] T The attitude angle error satisfies: The required roll and pitch angles are obtained by inverse kinematics of the position model: The rotational motion model then satisfies: Where, d Θ =[d x d y d z ] T u represents the unknown disturbance compensation amount. Θ =[u2 u3 u4] T This indicates the control input item.
4. The flight control method for tracking the trajectory of a coaxial dual-rotor UAV as described in claim 3, characterized in that, Based on the control objective, the attitude controller based on the adaptive radial basis function neural network (RBFNN) is configured as follows: For unknown disturbance quantities Provide compensation, take Use RBFNN to approximate the function f Θ =[f φ f θ f ψ ] T The uncertainty of the model is estimated and compensated in real time: Design an adaptive RBFNN network, defining the tracking error of the attitude system as follows: e=Θ d -I Define the auxiliary state error vector E as follows: Set an adaptive RBFNN to approximate the nonlinear unknown disturbance term f. Θ To obtain the output of the RBFNN network that satisfies: in, This represents the output vector of the RBFNN; This represents the input vector of the RBFNN, namely the variables such as the desired attitude angle and the actual value mentioned above. Here is the network weight coefficient matrix; J > 1 represents the number of hidden layer nodes in the network; H(·): For vector mapping, each function H j It is a Gaussian function; Where i is the number of input nodes in the network, j represents the j-th node in the hidden layer of the network, and c ij σ represents the position of the node. j w is the width of the j-th neuron. i * The optimal weight values set for the network; The output of the RBFNN network is: f Θ (I ij )=(W * ) T H(Θ in )+e Where ε is the neural network approximation error, satisfying ε max >0.
5. The flight control method for tracking the trajectory of a coaxial dual-rotor UAV as described in claim 4, characterized in that, According to the control objective, setting up an attitude controller based on an adaptive radial basis function neural network (RBFNN) further includes: Considering the horizontal motion model of CRA and the structure of RBFNN, design the control law: Design an adaptive algorithm to adjust the weights of the RBFNN network online, with the control law satisfying: Among them, K Θ It is a matrix of positive constants.
6. The flight control method for tracking the trajectory of a coaxial dual-rotor UAV as described in claim 4, characterized in that, Based on the motion characteristics of CRA trajectory tracking, the position controller based on adaptive parameter estimation is configured as follows: Based on the drone's location and desired location Determine the position error: Define an auxiliary function: Among them, Γ1=diag[λ1 λ2 λ3]; Speed error is defined as: Set position control input: in, For k p The estimated value, For d p The estimated value is Γ2 = diag[λ4 λ5 λ6]; The update rules for the estimated displacement resistance coefficient and the unknown external disturbance value in the design control input satisfy the following: in, 7. A coaxial dual-rotor unmanned aerial vehicle, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, which, when executed by the processor, implements the steps of the flight control method for tracking the trajectory of a coaxial dual-rotor unmanned aerial vehicle as described in any one of claims 1 to 6.