CN-121978907-A - Method for adaptively controlling scheduled time of passenger-wing separation variant gliding aircraft

CN121978907ACN 121978907 ACN121978907 ACN 121978907ACN-121978907-A

Abstract

The invention discloses a preset time self-adaptive control method of a passenger wing separation variant gliding aircraft, which comprises the following steps of firstly constructing a time scale function, effectively avoiding unlimited increase and singular problems of control quantity, secondly designing a preset time sliding mode surface based on the time scale function, secondly providing a neural network parameter identification strategy, converting a system identification problem into a parameter estimation problem so as to adapt to a flight scene with uncertain aerodynamic parameters, finally designing a preset time self-adaptive sliding mode controller, ensuring that the system state is converged in a specified time under the condition of no initial condition dependence, dynamically adjusting the self-adaptive gain of the controller according to the system environment, restraining jitter, reducing tracking errors, weakening the influence caused by concentrated uncertainty, and in conclusion, not only realizing accurate gesture tracking, but also achieving the preset time convergence characteristic independent of both initial conditions and control parameters, and having excellent quick response, stability and robustness.

Inventors

LIU LEI
LIU LIXIN
FAN HUIJIN
WANG BO

Assignees

华中科技大学

Dates

Publication Date: 20260505
Application Date: 20251222

Claims (7)

1. A method for adaptively controlling a passenger-wing separation variant gliding aircraft for a preset time is characterized by comprising the following steps: S1, constructing a pitching model and an attitude tracking error of a wing-in-wing separation variant gliding aircraft, and constructing a preset time self-adaptive controller according to the pitching model and the attitude tracking error; s1.1, constructing a pitching motion model of the wing-separated variant glider: , , , , , , Wherein, the method comprises the steps of, Is the angle of attack and the angle of attack, Is the angular velocity of the light beam, Is the deflection angle of the rudder, , , , , Is a function of the pneumatic parameters, , Is multi-source interference; s1.2, constructing a time scale function matched with the time scale function: , Wherein, the Is a positive constant which is a function of the current, Is a preset time specified by a user; s1.3, designing a preset time sliding mode surface according to the pitching motion model and the attitude tracking error, and constructing the preset time self-adaptive controller; S2, estimating uncertainty parameters in the preset time self-adaptive controller in real time through a neural network based on the state quantity of the aircraft; s2.1, constructing a neural network estimator for estimating lumped uncertainty in the pitching motion model on line; and S2.2, designing a weight self-adaptive law of the neural network estimator, and adjusting the neural network parameters on line to minimize estimation errors.
2. The method for adaptively controlling the predetermined time of the passenger-wing separation variant glider according to claim 1, wherein the step S1.3 comprises the following steps: S1.3.1 define attitude tracking error , Is a desired angle of attack command; S1.3.2 the preset time sliding mode surface is: Here, where And (2) and ; The derivative of the slip form surface is: , Wherein, the , S1.3.3 the set controller approach law structure is as follows: , S1.3.4 combining the derivative of the sliding mode surface with the approach law structure, setting the preset time self-adaptive controller to be: , Wherein, the And Using the neural network estimator to estimate in real time, Is an adaptive gain, and , 。
3. A method for adaptively controlling a passenger-wing separation variant glider for a predetermined time according to claim 2, wherein the method comprises the following steps of Setting the neural network estimator as: , , Wherein, the Is the angle of attack and the angle of attack, Is the angular velocity of the light beam, , Is a matrix of weights for the neural network, Is the number of nodes of the neural network, Is a neural network gaussian basis function.
4. A method of adaptive control of a wing-separated variant glider according to claim 3, wherein: the following form of gaussian basis function is satisfied: , Wherein, the Is the center vector of the vector, Is the width of the gaussian function.
5. The method for adaptively controlling the predetermined time of the passenger separation variant glider according to claim 4, wherein the weight adaptive law of the neural network estimator is as follows: , , Wherein, the , , , Is the gain parameter.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 5 when the program is executed by the processor.
7. A non-transitory computer readable storage medium, having stored thereon a computer program, characterized in that the program when executed by a processor realizes the steps of the method according to any of claims 1 to 5.

Description

Method for adaptively controlling scheduled time of passenger-wing separation variant gliding aircraft Technical Field The invention belongs to the field of control of variant aircrafts, and particularly relates to a preset time self-adaptive control method of a wing-separation variant gliding aircrafts. Background Because the traditional fixed-shape aircraft lacks the capability of self-adaptive deformation along with environmental changes, the pneumatic layout of each typical flight working condition is difficult to consider, and the survivability and the task efficiency are severely restricted. Thus, variant aircraft have been developed. Variant aircraft can autonomously change the profile to match the external environment, have wider flight envelope, better flight quality and stronger environmental adaptability than fixed profile aircraft, and are currently largely classified into variable-length variant aircraft, variable-chord variant aircraft, variable-thickness variant aircraft and wing-in-wing variant aircraft (refer to fig. 1). Compared to other variant aircraft, the winged-separated variant aircraft presents the following problems: (1) To achieve a varied flight mission, wing variants may be thrown away, resulting in changes in their dynamics and aerodynamic properties. The physical coupling between the fuselage and the wing causes a change in aerodynamic characteristics, manifested as uncertainty in aerodynamic parameters. Meanwhile, the flight attitude is required to be quickly adjusted in face of diversified flight tasks; (2) The flying environment facing the wing-in-wing separation variant aircraft is relatively complex, and involves task changes and wing variants, so that the flying system has concentrated uncertainty, including multi-source disturbance, unmodeled dynamics and the like; Thus, ensuring the rapidity, stability and robustness of attitude adjustment of a wing-in-wing separation variant aircraft during flight presents challenges. The design of the controller is critical to the safety, accurate performance and efficiency of the aircraft. There have been many studies on controller designs for aircraft such as optimal control, model predictive control, sliding mode control, and small gain control, which are capable of solving problems of attitude tracking control and stability on conventional fixed-profile aircraft, however, control performance in variant aircraft is unsatisfactory, and there is a need to design controllers for variant aircraft so that a wing-in-wing variant aircraft can have attitude control performance at a prescribed time and face a flight environment with uncertainty in aerodynamic parameters. Disclosure of Invention The invention aims to design a passenger-wing separation variant gliding aircraft, which is adaptive to complex environment and task requirements by changing aerodynamic configuration of the passenger-wing separation variant gliding aircraft through abandoned wings, and aims to solve the problems of dynamic characteristics, aerodynamic characteristic changes and uncertainty in lumped, and provides a passenger-wing separation variant gliding aircraft preset time self-adaptive control method based on neural network parameter identification. In order to achieve the above purpose, the invention adopts the following technical scheme steps: s1, designing a preset time self-adaptive controller of the aircraft according to a pitching model and an attitude tracking error of the aircraft with the variant wings; s2, according to the state quantity ,Is the angle of attack and the angle of attack,Is angular velocity, and constructs neural network parameter estimation for the controllerAnd: , , Wherein, the ,Is a matrix of weights for the neural network,Is the number of nodes of the neural network,Is a neural network gaussian basis function. Compared with the prior art, the invention has the following advantages: 1. The control design of the aircraft based on the wing-separated variant glider is that the traditional aircraft can only optimize the aerodynamic profile aiming at limited working conditions, and the wing-separated variant aircraft can obtain the optimal aerodynamic performance by adjusting variants in different flight phases, so that the wing-separated variant aircraft can take into account various task modes such as high-speed cruising, long-endurance glider, flexible maneuvering and the like. 2. The invention designs the neural network parameter identification, namely the pneumatic parameter uncertainty of the aircraft with the wing separation variant, the neural network parameter identification can convert the system identification problem into the parameter estimation problem so as to estimate the unknown weight of the network model, and the dependence on the prior knowledge of the accurate parameter is relieved based on the dynamic approximation of the neural network. 3. The invention designs the self-adaptive control method for the preset time,