Variant aircraft terminal sliding mode control method based on recurrent neural network observer
By using the observer and terminal sliding mode surface based on recurrent neural network in the variant aircraft, the problem of disturbance impact in the posture changes of the variant aircraft is solved, and more accurate disturbance estimation and fast response control effects are achieved.
Patent Information
- Application Number
- CN202410915969.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-09
AI Technical Summary
The additional disturbances generated by the variant aircraft during attitude changes seriously affect the stability of the aircraft attitude. The traditional disturbance estimation method relies on the current time information and cannot effectively track the rapidly changing aircraft state.
Using an observer based on a recurrent neural network, combined with the terminal sliding mode surface, a posture dynamic model of a variant aircraft is constructed, and a recurrent neural network observer is designed to estimate uncertain perturbation terms, and the control amount is obtained through the terminal sliding mode surface, using the past The measurement information at the moment allows for more accurate disturbance estimation and rapid response.
Improves the accuracy of disturbance estimation and control response speed, suitable for variant aircraft that quickly track changes, significantly improving the stability and efficiency of attitude control.
Smart Images

Figure CN118915435A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of aircraft control, and relates to a terminal sliding mode control method for a variant aircraft, and specifically to a terminal sliding mode control method for a variant aircraft based on a recurrent neural network observer. Background Art
[0002] As a new concept aircraft, the morphing aircraft can expand the flight envelope and improve flight performance by actively changing its shape. However, the additional disturbances generated by the morphing process seriously affect the attitude stability of the aircraft, and the state of the aircraft changes greatly before and after the morphing, so the attitude control system needs to achieve fast tracking. Summary of the invention
[0003] In order to solve the above problems existing in the background technology, the present invention provides a terminal sliding mode control method of a variant aircraft based on a recurrent neural network observer.
[0004] The objective of the present invention is achieved through the following technical solutions:
[0005] A terminal sliding mode control method for a variant aircraft based on a recurrent neural network observer comprises the following steps:
[0006] Step 1: Construct the attitude dynamics and kinematics model of the variant aircraft, and form an attitude control error model by subtracting it from the attitude command;
[0007] Step 2: Construct the terminal sliding surface based on the attitude control error model;
[0008] Step 3: Design an observer based on a recurrent neural network to observe the uncertain disturbance terms in the variant aircraft model;
[0009] Step 4: Based on the terminal sliding surface and the output value of the observer, the control quantity of the morphing aircraft is obtained.
[0010] Compared with the prior art, the present invention has the following advantages:
[0011] The traditional disturbance estimation method of the morphing aircraft only uses the measurement information at the current moment, while the recurrent neural network observer of the present invention can use the measurement information of the current moment and the past moment at the same time, and the estimation of the disturbance is more accurate. At the same time, the terminal sliding surface of the present invention can significantly improve the response speed, which is more suitable for objects such as morphing aircraft that need to be tracked quickly. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 This is a flow chart of the terminal sliding mode control method for a variant aircraft based on a recurrent neural network observer. DETAILED DESCRIPTION
[0013] The technical solution of the present invention is further described below in conjunction with the accompanying drawings, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention should be included in the protection scope of the present invention.
[0014] The present invention provides a terminal sliding mode control method for a variant aircraft based on a recurrent neural network observer, such as Figure 1 As shown, the method comprises the following steps:
[0015] Step 1: Construct the attitude dynamics and kinematics model of the variant aircraft, and form the attitude control error model by subtracting it from the attitude command. The specific steps are as follows:
[0016] Step 1.1: The attitude dynamics and kinematics model of the morphing aircraft is:
[0017]
[0018] Where: is the first-order derivative of Ω with respect to time; is the attitude angle vector of the aircraft; is the pitch angle; ψ is the yaw angle; γ is the roll angle; R is the attitude conversion matrix, is the first-order derivative of ω with respect to time; ω=[ω x ,ω y ,ω z ] is the attitude angular velocity vector of the aircraft; ω x is the rolling angular velocity; ω y is the yaw angular velocity; ω z is the pitch angular velocity; J represents the moment of inertia matrix of the aircraft; B1 represents the control torque coefficient matrix; δ=[δ x ,δ y ,δ z ] represents the control input; δ x is the deflection angle of the aileron; δ y is the deflection angle of the rudder; δ z is the deflection angle of the elevator; M SD Additional torque for variant; M SG is the additional gravity moment generated by the change of center of mass during the deformation process; M a is the aerodynamic torque; M d is the external disturbance torque.
[0019] Step 1.2: Set the aircraft attitude angle change command:
[0020] Ω c =[Ω cz ,Ω cy ,Ω cx ] (2)
[0021] Where: Ω cx is the roll channel command; Ω cy is the yaw channel command; Ω cz It is the pitch channel command.
[0022] The attitude angle tracking error is defined as:
[0023]
[0024] Where: is the pitch angle tracking error; x 1ψ is the yaw angle tracking error; x 1γ is the roll angle tracking error.
[0025] By continuously taking the second-order derivatives of equation (3), we can get the attitude control error model as follows:
[0026]
[0027] Where: represents the first-order derivative of x1 with respect to time; x2 represents the first-order derivative of the attitude angle tracking error; represents the second-order derivative of x1 with respect to time; u is the control variable, and u=RJ -1 B1δ; H is the uncertain disturbance term, and
[0028] Step 2: Construct the terminal sliding surface based on the attitude control error model.
[0029] The terminal sliding surface S is defined as:
[0030]
[0031] Where: λ1 and λ2 are positive gain coefficients; q and p are positive odd numbers, and
[0032] Step 3: Design an observer based on a recurrent neural network to observe the uncertain disturbance terms in the variant aircraft model. The specific steps are as follows:
[0033] Step 3.1 uses a three-layer recurrent neural network (RNN) to estimate the uncertain disturbance term H in real time. RNN consists of an input layer, a hidden layer, and an output layer. The RNN mapping is:
[0034]
[0035] Where: x i represents the input of RNN; y contains the input variables and output variables of RNN; N is the number of iterations; n is the number of nodes in the hidden layer; v k is the connection weight between the output layer and the hidden layer; sik and δ ik are the center and width of the radial basis function respectively; w i is the recursive weight coefficient of the output layer neuron; Φ k is the threshold of the kth neuron in the hidden layer, and m is the dimension of the input volume.
[0036] Step 3.2: To simplify the notation, define all parameter vectors of the RNN hidden layer δ, s, x and w, where: δ is the width of the hidden layer radial basis function δ ik The vector composed of s is the center of all the radial basis functions of the hidden layer s ik A vector composed of all input quantities x i A vector composed of , w is the recursive weight coefficient w of all output layer neurons i Vector composed of.
[0037] The output layer of the RNN is represented as a vector:
[0038] y(x,δ,s,w,v)=v T Φ(x,δ,s,w) (7)
[0039] where v = [v1 v2…v n ] T and Φ=[Φ1Φ2…Φ n ] T , v represents the weight of RNN, v i represents the i-th weight, Φ represents the activation function of RNN, Φ i Represents the i-th activation function, i=1,2,...,n.
[0040] Step 3.3: Define the RNN observer as:
[0041]
[0042] Where: is the estimated value of the disturbance term H by the RNN network; is the radial basis function estimate; is an estimate of the width of the radial basis function; is the estimate of the center of the radial basis function; is the estimated value of the recurrent weight coefficient of the output layer neurons.
[0043] Step 3.4: Design the adaptive law of the RNN observer as:
[0044]
[0045] Where: η1, η2, η3, η4 are the positive gain coefficients in the adaptive law; A1 is right The partial derivative matrix of right The partial derivative matrix of right The partial derivative matrix of .
[0046] Step 4: Based on the terminal sliding surface and the output value of the observer, the control quantity of the morphing aircraft is obtained.
[0047] The control quantity of the morphing aircraft is designed as follows:
[0048]
[0049] Where: Λ is a positive number, which is the approach gain.
Claims
1. A terminal sliding mode control method for a variant aircraft based on a recurrent neural network observer, characterized in that The method comprises the following steps: Step 1: Construct the attitude dynamics and kinematics model of the variant aircraft, and form an attitude control error model by subtracting it from the attitude command; Step 2: Construct the terminal sliding surface S based on the attitude control error model; Step 3: Design an observer based on a recurrent neural network to observe the uncertain disturbance terms in the variant aircraft model; Step 4: Based on the terminal sliding surface and the output value of the observer, the control quantity of the morphing aircraft is obtained.
2. The terminal sliding mode control method of a variant aircraft based on a recurrent neural network observer according to claim 1 is characterized in that The specific steps of step 1 are as follows: Step 1.1: The attitude dynamics and kinematics model of the morphing aircraft is: Where: is the first-order derivative of Ω with respect to time; is the attitude angle vector of the aircraft; is the pitch angle; ψ is the yaw angle; γ is the roll angle; R is the attitude conversion matrix, is the first-order derivative of ω with respect to time; ω=[ω x ,ω y ,ω z ] is the attitude angular velocity vector of the aircraft; ω x is the rolling angular velocity; ω y is the yaw angular velocity; ω z is the pitch angular velocity; J represents the moment of inertia matrix of the aircraft; B1 represents the control torque coefficient matrix; δ=[δ x ,δ y ,δ z ] represents the control input; δ x is the deflection angle of the aileron; δ y is the deflection angle of the rudder; δ z is the deflection angle of the elevator; M SD Additional torque for variant; M SG is the additional gravity moment generated by the change of center of mass during the deformation process; M a is the aerodynamic torque; M d is the external disturbance torque; Step 1.2: Set the aircraft attitude angle change command: Oh c =[Ω cz ,Oh cy ,Oh cx ] (2) Where: Ω cx is the roll channel command; Ω cy is the yaw channel command; Ω cz It is the pitch channel command; The attitude angle tracking error is defined as: Where: is the pitch angle tracking error; x 1ψ is the yaw angle tracking error; x 1γ is the rolling angle tracking error; By continuously taking the second-order derivatives of equation (3), we can get the attitude control error model as follows: Where: represents the first-order derivative of x1 with respect to time; x2 represents the first-order derivative of the attitude angle tracking error; represents the second-order derivative of x1 with respect to time; u is the control variable, and u=RJ -1 B1δ; H is the uncertain disturbance term, and 3. The terminal sliding mode control method of a variant aircraft based on a recurrent neural network observer according to claim 2 is characterized in that In step 2, the terminal sliding surface S is defined as: Where: λ1 and λ2 are positive gain coefficients; q and p are positive odd numbers, and 4. The terminal sliding mode control method of a variant aircraft based on a recurrent neural network observer according to claim 3 is characterized in that The specific steps of step 3 are as follows: Step 3.1 uses a three-layer RNN to estimate the uncertain disturbance term H in real time. The RNN consists of an input layer, a hidden layer, and an output layer. The RNN mapping is: Where: x i represents the input of RNN; y contains the input variables and output variables of RNN; N is the number of iterations; n is the number of nodes in the hidden layer; v k is the connection weight between the output layer and the hidden layer; s ik and δ ik are the center and width of the radial basis function respectively; w i is the recursive weight coefficient of the output layer neuron; Φ k is the threshold of the kth neuron in the hidden layer, and m is the dimension of the input; Step 3.2: To simplify the notation, define all parameter vectors of the RNN hidden layer δ, s, x and w, where: δ is the width of the hidden layer radial basis function δ ik The vector composed of s is the center of all the radial basis functions of the hidden layer s ik A vector composed of all input quantities x i A vector composed of , w is the recursive weight coefficient w of all output layer neurons i The vector composed of The output layer of the RNN is represented as a vector: y(x,δ,s,w,v)=v T Φ(x,δ,s,w) (7) where v = [v1v2…v n ] T and Φ=[Φ1Φ2...Φ n ] T , v represents the weight of RNN, v i represents the i-th weight, Φ represents the activation function of RNN, Φ i represents the i-th activation function, i=1,2,…,n; Step 3.3: Define the RNN observer as: Where: is the estimated value of the disturbance term H by the RNN network; is the radial basis function estimate; is an estimate of the width of the radial basis function; is the estimate of the center of the radial basis function; is the estimated value of the recursive weight coefficient of the output layer neuron; Step 3.4: Design the adaptive law of the RNN observer as: Where: η1, η2, η3, η4 are the positive gain coefficients in the adaptive law; A1 is right The partial derivative matrix of right The partial derivative matrix of right The partial derivative matrix of .
5. The terminal sliding mode control method of a variant aircraft based on a recurrent neural network observer according to claim 4 is characterized in that In step 4, the control amount of the variant aircraft is designed as follows: Where: Λ is a positive number, which is the approach gain.
Citation Information
Patent Citations
Attitude tracking control method for NSV (near-space vehicle)
CN108681331A
Aircraft attitude control method, system, medium and device
CN109343549A
Aircraft preset performance control method adaptive to large-amplitude maneuvering attitude
CN115390447A
Aircraft control system element gradient reinforcement learning design method
CN117369268A
Adaptive disturbance observation compensation aircraft attitude prediction control method
CN117666358A
