CN112068582B - A Tilting Rotor UAV Transition Mode Model Identification Method - Google Patents

A Tilting Rotor UAV Transition Mode Model Identification Method Download PDF

Info

Publication number
CN112068582B
CN112068582B CN202011016069.5A CN202011016069A CN112068582B CN 112068582 B CN112068582 B CN 112068582B CN 202011016069 A CN202011016069 A CN 202011016069A CN 112068582 B CN112068582 B CN 112068582B
Authority
CN
China
Prior art keywords
model
identification
control
test
moment
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN202011016069.5A
Other languages
Chinese (zh)
Other versions
CN112068582A (en
Inventor
蔡志浩
赵艳琪
赵江
王英勋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN202011016069.5A priority Critical patent/CN112068582B/en
Publication of CN112068582A publication Critical patent/CN112068582A/en
Application granted granted Critical
Publication of CN112068582B publication Critical patent/CN112068582B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
    • G05D1/0808—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The invention discloses a method for identifying a transition mode model of a tilt rotor unmanned aerial vehicle, which is characterized in that unknown parameters are determined by a frequency domain identification method, and the unmanned aerial vehicle under the conditions of different airspeeds and different nacelle inclination angles is identified through a test, so that the transition mode model of the tilt rotor unmanned aerial vehicle is obtained. According to the invention, the output quantity of the actuator is used as the input quantity of the model, and the control derivative of each actuator can be finally obtained through tests under any control distribution strategy of any controller, so that the established model not only can truly reflect the transition mode flight characteristics of the tilt rotor unmanned aerial vehicle, but also can provide convenience for the design of a control system.

Description

Method for identifying transition mode model of tilt rotor unmanned aerial vehicle
Technical Field
The invention belongs to the technical field of flight control and modeling of unmanned aerial vehicles, and particularly relates to a method for identifying a transition model of an unmanned aerial vehicle with a tilt rotor wing.
Background
The unmanned plane with the tilting rotor wing is a novel vertical take-off and landing aircraft with the characteristics of a fixed-wing aircraft and a helicopter, and is a hotspot of research at home and abroad since birth. Engine nacelles are respectively arranged on two sides of the wings of the unmanned aerial vehicle, and the taking-off and landing are realized similarly to a double-rotor helicopter; when flying forward, the nacelle tilts forward, and the rotor generates forward thrust, similar to a propeller-driven aircraft. The tilt rotor unmanned aerial vehicle has the vertical take-off and landing and hovering capabilities of a conventional helicopter, has the advantages of high speed, large load capacity, long range and the like of a fixed-wing aircraft, and has high application value for military use and civil use.
The tilt rotor unmanned aerial vehicle has three flight modes, namely a helicopter mode, a transition mode and a fixed wing mode. In which the transition mode is controlled by both the rotor and the rudder, the derivative of the rotor and rudder control is not only related to the nacelle pitch and airspeed, but also changes as the controller or control distribution method changes, and both sets of control systems are more complex than if only one set of control systems were used. There is therefore a need for an efficient method of modeling, through reasonable experimentation, the actuator control derivatives required for control system design under any controller-any control distribution strategy.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides a method for identifying a transition mode model of a tilt rotor unmanned aerial vehicle, which utilizes CIFER software to identify a multi-input multi-output unmanned aerial vehicle system. The specific technical scheme of the invention is as follows:
the method for identifying the transition mode model of the tilt rotor unmanned aerial vehicle is characterized by comprising the following steps of:
s1: establishing a nonlinear model of the tilt rotor unmanned aerial vehicle according to the principles of flight mechanics, aerodynamic mechanics and kinematics;
s2: designing a basic control law, a control distribution strategy and an identification scheme aiming at the transition mode flight characteristics of the tilt rotor unmanned aerial vehicle;
s3: testing and collecting test data, selecting data related to the parameters to be identified from the test data, preprocessing the data, and converting the preprocessed data into a data type which can be processed by CIFER software;
s4: performing small-disturbance linearization on the nonlinear model obtained in the step S1, establishing a transfer function and a state equation, and determining a linear model for model parameter identification;
s5: the data obtained in the step S3 is processed by FRESPID, MISOSA and COMPOSITE parts of CIFER software to generate a MIMO frequency response database, and coherence values are reserved
Figure BDA0002699108130000021
And the data not oscillating with the coherence value curve is used for identification, if the frequency domain data in the identification frequency band exceeds two thirds and does not accord with the coherence value standard set by the CIFER software, the step S2 is returned to redesign the identification scheme;
s6: performing frequency domain identification through the parameterized model identification functions of the CIFER software, namely NAVFIT and DERIVD, and meeting JaveThe parameter to be identified is obtained according to the criterion of less than or equal to 100, JaveIn order to identify the cost function, the linear model obtained in the step S4 is subjected to time domain verification through VERIFY, and if rms fitting error J occursrmsAnd (3) the time domain verification of the identified model is represented when the coefficient TIC of the Theil inequality is less than or equal to 1.0-2.0 and less than or equal to 0.25-0.3, otherwise, the step S4 is returned, the linear model is determined again, and the complete model of the transition mode of the tilt rotor unmanned aerial vehicle is obtained through identification of each channel.
Further, the basic control law and the control distribution strategy designed in step S2 are implemented by matching an INDI controller with a pseudo-inverse method to control an attitude angle, and are specifically implemented as follows:
the nonlinear dynamical system equation of the unmanned aerial vehicle is as follows:
Figure BDA0002699108130000022
wherein x is an n-dimensional state quantity,
Figure BDA0002699108130000023
is the derivative of x, u is the p-dimensional control input; y is an m-dimensional output, f (x) is a non-linear function of x, h (x) is a linear function of x, and g (x, u) is a linear function of x and u;
definition of
Figure BDA0002699108130000024
Wherein A is partial derivative of f (x) + g (x, u) to x, B is partial derivative of g (x, u) to u, and the state quantity and the input quantity at a certain moment are (x)0,u0) The desired closed loop dynamics are
Figure BDA0002699108130000025
When the calculation frequency of the control law is high enough and the step length is small enough, the incremental dynamic inverse control law is designed as follows:
Figure BDA0002699108130000026
where Δ u is the control increment, hxIs the partial derivative of g (x, u) with respect to x.
Further, the identification scheme designed in step S2 includes test subjects, test times, input signals and identification methods; the identification scheme specifically comprises the following steps:
in each test, the identification tests of the pitching channel, the rolling channel and the yawing channel are respectively carried out two to three times under the condition that the airspeed and the nacelle inclination angle are controlled to be kept unchanged as much as possible, the channel to be identified is called a main channel, and the rest channels are called auxiliary channels. Because the transition mode has two sets of control systems, the hybrid control mode can only identify the input-output relationship between the hybrid control quantity and the response of the unmanned aerial vehicle, and in order to obtain the control derivative of the actuator, a test only with the action of a rotor wing or only with the action of a control plane is required. Limited by test conditions and control system design, test data of the unmanned aerial vehicle under the conditions of rotor operation and control surface operation are not easy to obtain at the same time, at the moment, a mixed operation test is firstly carried out, and then a single operation mode capable of obtaining high-quality response is selected for testing.
The sweep frequency signal is used as the input signal of the main channel, the aircraft angle response caused by the sweep frequency amplitude is +/-5 degrees to +/-15 degrees, and the frequency range is 0.5 omegaBW≤ω≤2.5ω180，ωBWIs the bandwidth frequency, omega180The frequency corresponding to the phase angle of-180 degrees. The frequency sweep experiment is started after the tilt rotor unmanned aerial vehicle is in a balancing state for 3-5 seconds, the experiment is ended after the frequency sweep is ended and the tilt rotor unmanned aerial vehicle stays in the balancing state for at least 3 seconds, and the time of each experiment is recommended
Figure BDA0002699108130000031
ωminIs the minimum frequency, ωmin＝0.5ωBW。
And inputting small-amplitude white noise into the secondary channel to avoid the coupling of the input quantity, and adopting a direct identification method to identify, namely directly measuring the output quantity of the actuator and the output quantity of the model to identify.
Further, in the step S3, if the test is limited by the condition, only one single maneuver test is completed, and the other single maneuver response data is obtained by subtracting the completed single maneuver response data from the hybrid maneuver response data under the same condition.
Further, the longitudinal and lateral state equations established in step S4 are:
Figure BDA0002699108130000032
Figure BDA0002699108130000041
u, w, q, theta, v, p, r and phi are state quantities, u, v and w are disturbance speeds in the directions of x, y and z axes under a body axis system respectively, p, q and r are disturbance rolling angle speed, pitching angle speed and yaw angle speed respectively, and theta and phi are disturbance pitch angle and rolling angle respectively;
Figure BDA0002699108130000042
δe,δa,δrthe control variables are respectively a rotor wing longitudinal periodic variable pitch, a rotor wing total pitch differential, a rotor wing longitudinal periodic variable pitch differential, an elevator deflection angle, an aileron deflection angle and a rudder deflection angle; xu、Zu、MuPartial derivatives of the longitudinal force, lateral force and pitching moment, respectively, on u, Xw、Zw、MwPartial derivatives of the longitudinal force, lateral force and pitching moment, respectively, with respect to w, Xq、Zq、MqThe partial derivatives of the longitudinal force, lateral force and pitching moment on q,
Figure BDA0002699108130000043
respectively longitudinal force, lateral force and pitching moment
Figure BDA0002699108130000044
The partial derivative of (a) of (b),
Figure BDA0002699108130000045
longitudinal force, lateral force and pitching moment pairs delta respectivelyePartial derivatives of (d); y isv、Lv、NvPartial derivatives of the transverse force, the roll moment and the yaw moment, respectively, with respect to v, Yp、Lp、NpPartial derivatives of p, Y, of the transverse force, roll moment and yaw moment, respectivelyr、Lr、NrRespectively the partial derivatives of the lateral force, the roll moment and the yaw moment to r,
Figure BDA0002699108130000046
respectively, transverse force, rolling moment and yawing moment
Figure BDA0002699108130000047
The partial derivative of (a) of (b),
Figure BDA0002699108130000048
respectively, transverse force, rolling moment and yawing moment
Figure BDA0002699108130000049
The partial derivative of (a) of (b),
Figure BDA00026991081300000410
transverse force, roll moment and yaw moment respectivelyaThe partial derivative of (a) of (b),
Figure BDA00026991081300000411
transverse force, roll moment and yaw moment respectivelyrPartial derivatives of (d); u shape0、W0、Θ0Respectively the speed of the lower body axis in the x-axis direction in the trim state,The speed in the z-axis direction and the pitch angle under the body axis system; g represents the gravitational acceleration.
Further, the specific process of step S6 is as follows:
s6-1: fitting the frequency response pair in the MIMO frequency domain database obtained in the step S5 into a SISO transfer function model form through a NAVFIT part of CIFER software, and converting the fitting result of the transfer function into a state equation form to obtain an initial value of a state equation parameter;
s6-2: fitting the MIMO frequency domain data into a state equation through a DERIVD part of the CIFER software;
if the fitting condition of the identification result and the frequency response curve is not good, returning to the step S2 to re-determine the linear model identified by the model parameters, performing time domain verification on the linear model identified by the model parameters with determined parameters in a VERIFY part by adopting bipolar square waves, if the verification fails, proving that the identified model does not have model prediction capability, and then returning to the step S4 to re-determine the linear model identified by the model parameters.
And repeating the step S6 until all the channel actuator control derivatives are determined, so as to obtain a complete transition mode model of the tilt rotor unmanned aerial vehicle.
The invention has the beneficial effects that:
1. the method obtains the transition mode model of the tilt rotor unmanned aerial vehicle through the CIFER software identification result based on the test data, and is different from other models in that the output quantity of an actuator is used as the input quantity of the model instead of the mixed operation quantity.
2. The identification method can finally obtain the control derivative of each actuator through a test under any control distribution strategy of any controller, so that the established model not only can truly reflect the transition mode flight characteristics of the tilt rotor unmanned aerial vehicle, but also can provide convenience for the design of a control system.
Drawings
In order to illustrate embodiments of the present invention or technical solutions in the prior art more clearly, the drawings which are needed in the embodiments will be briefly described below, so that the features and advantages of the present invention can be understood more clearly by referring to the drawings, which are schematic and should not be construed as limiting the present invention in any way, and for a person skilled in the art, other drawings can be obtained on the basis of these drawings without any inventive effort. Wherein:
fig. 1 is a flow chart of a method for identifying a transition mode model of a tilt rotor unmanned aerial vehicle according to the present invention;
FIG. 2 is a schematic diagram of a basic controller according to an embodiment of the present invention;
FIGS. 3(a) - (f) are graphs of the roll channel frequency response fit results of the present invention;
FIGS. 4(a) - (f) are graphs of collective differential steering response fit results of the present invention;
FIGS. 5(a) - (b) are graphs of the rolling channel time domain verification results of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments of the present invention and features of the embodiments may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
As shown in fig. 1, the present invention provides a method for identifying the steering derivative of any actuator under any control allocation strategy of any controller, and particularly provides a method for identifying a transition model of a tilt rotor unmanned aerial vehicle. According to the method, unknown parameters are determined by a frequency domain identification method, and unmanned aerial vehicles with different airspeeds and under different nacelle inclination angles are identified through tests, so that a transition mode model of the tilt rotor unmanned aerial vehicle is obtained.
For the convenience of understanding the above technical aspects of the present invention, the following detailed description will be given of the above technical aspects of the present invention by way of specific examples.
Example 1
A method for identifying a transition mode model of a tilt rotor unmanned aerial vehicle comprises the following steps:
s1: according to the principles of flight mechanics, aerodynamic mechanics and kinematics, the Simulink is utilized to adopt a block combination mode and consider pneumatic coupling with large influence, mechanism modeling is carried out on the tilt rotor unmanned aerial vehicle, and a nonlinear model of the tilt rotor unmanned aerial vehicle is obtained.
The mechanism modeling adopts a modeling method which is a combination of block modeling and combined modeling, namely, the researched tilt rotor unmanned aerial vehicle is divided into a plurality of parts such as a fuselage, a wing, a stabilizing surface, an engine and a rotor wing, the aerodynamic force and the moment of each part are respectively calculated and superposed, and then the movement is calculated to obtain the movement state of the aircraft. And (3) in consideration of mutual coupling influence and model complexity among all the components, carrying out principle-based modeling on some coupling pneumatic effects which have the largest influence, and neglecting factors with smaller influence.
S2: basic control law and control distribution strategy and identification scheme are designed according to the transition mode flight characteristics of the tilt rotor unmanned aerial vehicle. The designed identification scheme comprises test subjects, test times, input signals and an identification method.
Considering the instability of the tilt rotor unmanned aerial vehicle, the controller must be designed to perform a closed loop test on the basis of the relatively coarse model obtained in step S1. Because the tilt rotor model has an unmodeled part and an uncertain part, which causes a large difference between the model and the actual model, the INDI control method has good robustness, has small dependence on the model and can perform nonlinear control, and therefore the INDI controller is adopted to cooperate with the pseudo-inverse method to control the attitude angle in the embodiment, as shown in fig. 2.
The method comprises the steps of carrying out an identification test aiming at a rolling channel under the condition of a given nacelle inclination angle and airspeed, waiting for 3s after balancing, inputting a sweep frequency signal 15s with an amplitude of 15 degrees and a frequency band of 1-30 rad/s to a rolling angle instruction, inputting white noise with a small amplitude to a channel which is not swept to avoid singularity, and waiting for 3s again to finish the test in a balancing state after the sweep frequency is finished. The simulation was performed twice in the hybrid steering state and twice in the rotor-only steering state.
S3: and carrying out simulation test, acquiring test data, selecting data related to the parameter to be identified, preprocessing the data, and converting the preprocessed data into a data type which can be processed by CIFER software.
Simulating and reading data through matlab, and selecting input quantity and observed quantity-related to the parameter to be identified from the data
Figure BDA0002699108130000071
δaP and phi, intercepting the sweep frequency data of the quantities in a trim state, abandoning experimental data obviously having errors, defects or poor quality, and converting the experimental data into a data type (dat file) capable of being processed by the CIFER software.
If the test is limited by the condition, only one single manipulation test is completed, and the other single manipulation response data is obtained by subtracting the single manipulation response data from the mixed manipulation response data under the same condition, and then the data is converted into the data type which can be processed by the CIFER software.
S4: and (5) carrying out small-disturbance linearization on the nonlinear model obtained in the step (S1), establishing a transfer function and a state equation, and determining a linear model for model parameter identification. Because some of the stable derivatives and the steering derivative lines are unknown or inaccurate after the model is modeled probabilistically, the parameters are the parameters to be identified. The equation of state of the roll channel in the transition mode is as follows:
Figure BDA0002699108130000072
wherein, p and phi are state quantities and respectively represent the rolling angle and the rolling angular speed;
Figure BDA0002699108130000073
δathe total distance differential of the rotor wing and the deflection angle of the aileron are respectively expressed as control quantity; l isp、
Figure BDA0002699108130000074
Respectively represent the roll torque pairs p,
Figure BDA0002699108130000075
δaPartial derivatives of (a).
S5: the data obtained in step S3 is processed by FRESPID, MISOSA, COMPOSITE and other parts of CIFER software to generate MIMO frequency response database, and coherence value is reserved
Figure BDA0002699108130000076
And the data with smooth coherence value curve is not oscillated for identification, the identification frequency band is 1.26-30 rad/S in the embodiment, if the coherence value of the data with more than two thirds of frequency domain data in the identification frequency band does not accord with the coherence value standard set by the CIFER software, the step S2 is returned to redesign the identification scheme, otherwise, the step S6 is continued;
if the frequency band identification or the sweep frequency amplitude is not reasonable to select, or the test flight condition is not good, the nonlinear link or the large noise is caused, the analysis is the problem of test flight, the step S3 is returned to perform the test, and if the problem is the problem of scheme design, the step S2 is returned to.
S6: performing frequency domain identification through a parameterized model identification function of the CIFER software, namely NAVFIT and DERIVD, obtaining parameters to be identified, and setting the state quantity of a rolling channel under the control of a rotor wing to be
Figure BDA0002699108130000077
Figure BDA0002699108130000078
For roll rate under differential collective pitch steering,
Figure BDA0002699108130000079
identifying a cost function J for a roll angle under differential collective pitch control, under rotor controlave40.6 < 100. Bipolar square wave frequency 0.5Hz, partial time domain verification TIC 0.068 by VERIFY<0.25，Jrms0.0886 < 1, wherein TIC refers to the value of Theil inequality coefficient, JrmsRepresents the rms fit error and generally reflects an acceptable level of accuracy in the time domain of flight dynamics modeling. Identifying the obtained formThe equation of state is:
Figure BDA0002699108130000081
the state quantity of the rolling channel under the control of the control surface is set as
Figure BDA0002699108130000082
Figure BDA0002699108130000083
Is the roll angular velocity under the control of the control surface,
Figure BDA0002699108130000084
the rolling angle under the control surface manipulation is the state quantity of a rolling channel under the mixed manipulation, namely x ═ p phi]TThe state quantity caused by the operation of the rotor under the mixed-mode operation is
Figure BDA0002699108130000085
Because simulation only obtains
Figure BDA0002699108130000086
And x, so returning to step S3 the control surface manipulation data is calculated by:
Figure BDA0002699108130000087
sequentially determining a parameter identification model, generating a corresponding frequency response library, and performing hybrid manipulation identification to obtain a state equation as follows:
Figure BDA0002699108130000088
the roll channel frequency aileron steering response fitting results are shown in FIG. 3, Jave47.7 < 100, the fitting results of the collective differential steering response are shown in fig. 4(a) - (f), wherein fig. 4(a) - (b) represent logarithmic amplitude-frequency curves, fig. 4(c) - (d) represent logarithmic phase-frequency curves, and fig. 4(e) - (f)) For the correlation value curve, the solid line represents the simulated data curve and the dashed line represents the model fitting curve. Bipolar square wave frequency 1Hz, two indices verified by VERIFY partial time domain: j. the design is a squarerms＝0.107＜1，TIC＝0.099<0.25, the time domain verification results all meet the standard are shown in fig. 5, the solid line is a flight data time domain response curve, and the dotted line is a model time domain response curve.
The essence of parameter fitting is the optimization process of the optimization algorithm, so when the number of parameters to be identified is large or the optimization is so large that the range cannot be determined, the initial values of some parameters need to be determined first, and then the parameter identification of the state equation needs to be performed, so the NAVFIT part is performed first, and then the DERIVID part is performed.
And if the precision does not meet the requirement, returning to the step S4, re-determining the linear model, and finally obtaining a complete model of the transition mode of the tilt rotor unmanned aerial vehicle by identifying each channel.
The method obtains the linear model of the transition mode of the tilt rotor unmanned aerial vehicle through the CIFER software identification result based on the unmanned aerial vehicle test data, and is different from the traditional modeling in that the control derivative of each actuator is obtained instead of the control derivative of the mixed control quantity, so that the method is of great significance for understanding the flight characteristics of the transition mode of the tilt rotor unmanned aerial vehicle and further designing a control system. By using the method, the actuator control derivative under any control distribution strategy of any controller can be obtained, and the finally obtained unmanned aerial vehicle linear model can reflect the flight response of the tilt rotor unmanned aerial vehicle under a certain precision requirement.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (4)

1.一种倾转旋翼无人机过渡模式模型辨识方法，其特征在于，包括以下步骤：1. a tilt-rotor unmanned aerial vehicle transition mode model identification method, is characterized in that, comprises the following steps: S1：根据飞行力学、空力动力学及运动学原理，建立倾转旋翼无人机的非线性模型；S1: According to the principles of flight mechanics, aerodynamics and kinematics, establish a nonlinear model of the tilt-rotor UAV; S2：针对倾转旋翼无人机过渡模式飞行特性设计基本控制律和控制分配策略以及辨识方案；S2: Design the basic control law, control distribution strategy and identification scheme for the flight characteristics of the tilt-rotor UAV in transition mode; 辨识方案包括试验科目、试验次数、输入信号和辨识方法；辨识方案具体为：每次试验应在控制空速和短舱倾角尽量保持不变的情况下分别进行俯仰、滚转及偏航通道的辨识试验两到三次，待辨识通道称为主通道，其余称为副通道；由于过渡模式存在两套操纵系统，采用混合操纵方式仅能辨识出混合操纵量与无人机响应的输入输出关系，为得到作动器操纵导数，还需要进行仅有旋翼作用或者仅有舵面作用的试验；受试验条件以及控制系统设计所限，一般不容易同时得到无人机仅在旋翼操纵和仅在舵面操纵情况下的试验数据，此时先进行混合操纵的试验，再选取可以获得高质量响应的单操纵方式进行试验；The identification scheme includes test subjects, number of tests, input signals and identification methods; the identification scheme is specific as follows: each test should conduct pitch, roll and yaw channels respectively under the condition that the controlled airspeed and nacelle inclination remain unchanged as much as possible. The identification test is carried out two or three times. The channel to be identified is called the main channel, and the rest are called the auxiliary channel. Since there are two sets of control systems in the transition mode, the mixed control method can only identify the input and output relationship between the mixed control amount and the UAV response. In order to obtain the control derivative of the actuator, it is also necessary to conduct a test with only the rotor action or only the rudder action; limited by the test conditions and the design of the control system, it is generally not easy to obtain the UAV only in the rotor control and only in the rudder at the same time. The test data in the case of surface manipulation, at this time, the mixed manipulation test is performed first, and then the single manipulation mode that can obtain high-quality response is selected for the test; 采用扫频信作为主通道的输入信号，扫频幅值引起的飞机角度响应为±5°～±15°，频率范围为0.5ωBW≤ω≤2.5ω180，ωBW为带宽频率，ω180为相角-180度时所对应的频率；扫频实验在倾转旋翼无人机处于配平状态3至5秒后开始，结束扫频后在配平状态停留至少3秒后结束一次试验，每次试验时间推荐
Figure FDA0003280901160000011
ωmin为最小频率，ωmin＝0.5ωBW；The frequency sweep signal is used as the input signal of the main channel, the angle response of the aircraft caused by the frequency sweep amplitude is ±5°～±15°, and the frequency range is 0.5ω BW ≤ω≤2.5ω 180 , ω BW is the bandwidth frequency, ω 180 It is the frequency corresponding to the phase angle of -180 degrees; the frequency sweep experiment starts after the tiltrotor UAV is in the trim state for 3 to 5 seconds, and after the frequency sweep is completed, the experiment ends after staying in the trim state for at least 3 seconds. Test time recommendation
Figure FDA0003280901160000011
ω min is the minimum frequency, ω min =0.5ω BW ; 在副通道输入小幅值白噪声避免产生输入量的耦合，采用直接辨识方法进行辨识，即直接测量作动器输出量和模型输出量进行辨识；Input small-amplitude white noise in the secondary channel to avoid coupling of input quantities, and use the direct identification method for identification, that is, directly measure the output of the actuator and the output of the model for identification; S3：试验并采集试验数据，从中选取与待辨识参数有关的数据，对数据进行预处理之后转换成CIFER软件能够处理的数据类型；S3: Test and collect test data, select data related to the parameters to be identified, and convert the data into a data type that can be processed by CIFER software after preprocessing; S4：对步骤S1得到的非线性模型进行小扰动线性化，建立传递函数和状态方程，确定模型参数辨识的线性模型；其中，建立的纵向与横侧向状态方程为：S4: perform small disturbance linearization on the nonlinear model obtained in step S1, establish a transfer function and a state equation, and determine a linear model for model parameter identification; wherein, the established longitudinal and lateral state equations are:
Figure FDA0003280901160000012
Figure FDA0003280901160000012
Figure FDA0003280901160000013
Figure FDA0003280901160000013
 其中，u，w，q，θ，v，p，r，φ为状态量，u、v、w分别为体轴系下x、y、z轴方向扰动速度，p、q、r分别为扰动滚转角速度、俯仰角速度和偏航角速度，θ、φ分别扰动俯仰角和滚转角；
Figure FDA0003280901160000021
δe,δa,δr为控制量，分别为旋翼纵向周期变距、旋翼总距差动、旋翼纵向周期变距差动、升降舵偏转角、副翼偏转角和方向舵偏转角；Xu、Zu、Mu分别为纵向力、侧向力和俯仰力矩对u的偏导数，Xw、Zw、Mw分别为纵向力、侧向力和俯仰力矩对w的偏导数，Xq、Zq、Mq分别为纵向力、侧向力和俯仰力矩对q的偏导数，
Figure FDA0003280901160000022
分别为纵向力、侧向力和俯仰力矩对
Figure FDA0003280901160000023
的偏导数，
Figure FDA0003280901160000024
分别为纵向力、侧向力和俯仰力矩对δe的偏导数；Yv、Lv、Nv分别为横向力、滚转力矩和偏航力矩对v的偏导数,Yp、Lp、Np分别为横向力、滚转力矩和偏航力矩对p的偏导数，Yr、Lr、Nr分别为横向力、滚转力矩和偏航力矩对r的偏导数，
Figure FDA0003280901160000025
分别为横向力、滚转力矩和偏航力矩对
Figure FDA0003280901160000026
的偏导数，
Figure FDA0003280901160000027
分别为横向力、滚转力矩和偏航力矩对
Figure FDA0003280901160000028
的偏导数，
Figure FDA0003280901160000029
分别为横向力、滚转力矩和偏航力矩对δa的偏导数，
Figure FDA00032809011600000210
分别为横向力、滚转力矩和偏航力矩对δr的偏导数；U0、W0、Θ0分别为配平状态下体轴系下x轴方向速度、体轴系下z轴方向速度和俯仰角；g表示重力加速度；Among them, u, w, q, θ, v, p, r, φ are the state quantities, u, v, and w are the perturbation speeds of the x, y, and z axes in the body axis system, respectively, and p, q, and r are the perturbation speeds, respectively. Roll angular velocity, pitch angular velocity and yaw angular velocity, θ, φ disturb pitch angle and roll angle respectively;
Figure FDA0003280901160000021
δ e , δ a , δ r are the control variables, which are respectively the rotor longitudinal cyclic pitch, rotor collective pitch differential, rotor longitudinal cyclic pitch differential, elevator deflection angle, aileron deflection angle and rudder deflection angle; X u , Z u , Mu are the partial derivatives of longitudinal force, lateral force and pitching moment to u respectively, Xw , Zw , Mw are the partial derivatives of longitudinal force, lateral force and pitching moment to w respectively, X q , Z q , M q are the partial derivatives of longitudinal force, lateral force and pitching moment to q, respectively,
Figure FDA0003280901160000022
are the pair of longitudinal force, lateral force and pitching moment, respectively
Figure FDA0003280901160000023
The partial derivative of ,
Figure FDA0003280901160000024
are the partial derivatives of longitudinal force, lateral force and pitching moment to δ e respectively; Y v , L v , and N v are the partial derivatives of lateral force, rolling moment and yaw moment to v respectively, Y p , L p , N p are the partial derivatives of lateral force, rolling moment and yaw moment with respect to p, respectively, Y r , L r , and N r are the partial derivatives of lateral force, rolling moment and yaw moment with respect to r, respectively,
Figure FDA0003280901160000025
are the lateral force, roll moment and yaw moment pair, respectively
Figure FDA0003280901160000026
The partial derivative of ,
Figure FDA0003280901160000027
are the lateral force, roll moment and yaw moment pair, respectively
Figure FDA0003280901160000028
The partial derivative of ,
Figure FDA0003280901160000029
are the partial derivatives of lateral force, rolling moment and yaw moment with respect to δ a , respectively,
Figure FDA00032809011600000210
are the partial derivatives of lateral force, rolling moment and yaw moment to δ r , respectively; U 0 , W 0 , Θ 0 are the velocity in the x-axis direction under the body shafting, the velocity in the z-axis direction under the body shafting, and the pitch in the trim state, respectively angle; g represents the acceleration of gravity; S5：将步骤S3得到的数据经过CIFER软件的FRESPID、MISOSA和COMPOSITE部分生成MIMO频率响应数据库，保留相干值
Figure FDA00032809011600000211
且相干值曲线不振荡的数据用于辨识，若辨识频段内的频域数据超过三分之二不符合CIFER软件设定的相干值标准，则返回步骤S2重新设计辨识方案；S5: The data obtained in step S3 is passed through the FRESPID, MISOSA and COMPOSITE parts of the CIFER software to generate a MIMO frequency response database, and the coherence value is retained
Figure FDA00032809011600000211
And the data whose coherence value curve does not oscillate is used for identification. If more than two-thirds of the frequency domain data in the identification frequency band do not meet the coherence value standard set by the CIFER software, then return to step S2 to redesign the identification scheme; S6：通过CIFER软件的参数化模型辨识功能即NAVFIT和DERIVID进行频域辨识，满足Jave≤100准则得到待辨识参数，Jave为辨识代价函数，再将步骤S4得到的线性模型通过VERIFY进行时域验证，若rms拟合误差Jrms≤1.0～2.0且Theil不等式系数TIC≤0.25～0.3则表示辨识得到的模型通过时域验证，否则，返回步骤S4，重新确定线性模型，通过对各个通道的辨识，得到完整的倾转旋翼无人机过渡模式的模型。S6: Perform frequency domain identification through the parameterized model identification function of CIFER software, namely NAVFIT and DERIVID, and obtain the parameters to be identified by satisfying the criterion of J ave ≤ 100, where J ave is the identification cost function, and then pass the linear model obtained in step S4 through VERIFY. Domain verification, if the rms fitting error J rms ≤1.0～2.0 and Theil inequality coefficient TIC≤0.25～0.3, it means that the identified model passes the time domain verification, otherwise, go back to step S4, re-determine the linear model, Identify and obtain a complete model of the transition mode of the tilt-rotor UAV. 2.根据权利要求1所述的一种倾转旋翼无人机过渡模式模型辨识方法，其特征在于，所述步骤S2设计的基本控制律及控制分配策略为INDI控制器配合伪逆法进行姿态角的控制，具体实现如下：2. a kind of tilt-rotor unmanned aerial vehicle transition mode model identification method according to claim 1, is characterized in that, the basic control law and control distribution strategy of described step S2 design is that INDI controller cooperates with pseudo-inverse method to carry out attitude The control of the angle is implemented as follows: 无人机非线性动力学系统方程为：The nonlinear dynamic system equation of the UAV is:
Figure FDA0003280901160000031
Figure FDA0003280901160000031
 其中，x为n维状态量，
Figure FDA0003280901160000032
为x的导数，u为p维控制输入；y为m维输出，f(x)为x的非线性函数，h(x)为x的线性函数，g(x,u)为x和u线性函数；Among them, x is the n-dimensional state quantity,
Figure FDA0003280901160000032
is the derivative of x, u is the p-dimensional control input; y is the m-dimensional output, f(x) is the nonlinear function of x, h(x) is the linear function of x, and g(x, u) is the linear function of x and u function; 定义definition
Figure FDA0003280901160000033
Figure FDA0003280901160000033
 其中，A为f(x)+g(x,u)对x的偏导数，B为g(x,u)对u的偏导数，设某一时刻状态量和输入量为(x0,u0)，期望的闭环动态特性为
Figure FDA0003280901160000034
当控制律的计算频率足够高，步长足够小时，设计增量动态逆控制律为：Among them, A is the partial derivative of f(x)+g(x, u) to x, B is the partial derivative of g(x, u) to u, and the state quantity and input quantity at a certain moment are (x 0 , u 0 ), the desired closed-loop dynamic characteristics are
Figure FDA0003280901160000034
When the calculation frequency of the control law is high enough and the step size is small enough, the design incremental dynamic inverse control law is:
Figure FDA0003280901160000035
Figure FDA0003280901160000035
 其中，Δu为控制增量，hx为g(x,u)对x的偏导数。Among them, Δu is the control increment, and h x is the partial derivative of g(x, u) with respect to x. 3.根据权利要求1或2所述的一种倾转旋翼无人机过渡模式模型辨识方法，其特征在于，所述步骤S3中，如果试验受条件所限只完成了一种单操纵试验，另一种单操纵响应数据则通过在同一条件下的混合操纵响应数据减去已完成的单操纵响应数据得到。3. a kind of tilt-rotor unmanned aerial vehicle transition mode model identification method according to claim 1 and 2, is characterized in that, in described step S3, if the test is limited by condition, only completed a kind of single manipulation test, Another single manipulation response data is obtained by subtracting the completed single manipulation response data from the mixed manipulation response data under the same conditions. 4.根据权利要求1或2所述的一种倾转旋翼无人机过渡模式模型辨识方法，其特征在于，所述步骤S6的具体过程为：4. a kind of tilt-rotor unmanned aerial vehicle transition mode model identification method according to claim 1 and 2, is characterized in that, the concrete process of described step S6 is: S6-1：通过CIFER软件的NAVFIT部分，将步骤S5得到的MIMO频域数据库中的频率响应对拟合为SISO传递函数模型形式，再将传递函数拟合结果转化为状态方程的形式，得到状态方程参数初值；S6-1: Through the NAVFIT part of the CIFER software, fit the frequency response pair in the MIMO frequency domain database obtained in step S5 into the SISO transfer function model form, and then convert the transfer function fitting result into the form of the state equation to obtain the state The initial value of the equation parameters; S6-2：通过CIFER软件的DERIVID部分，将MIMO频域数据拟合成状态方程；S6-2: Fit the MIMO frequency domain data into a state equation through the DERIVID part of the CIFER software; 若辨识结果与频率响应曲线拟合情况不佳，则返回步骤S2重新确定模型参数辨识的线性模型，采用双极方波在VERIFY部分对参数已确定的模型参数辨识的线性模型进行时域验证，验证不通过则证明辨识得到的模型不具备模型预测能力，之后返回步骤S4重新确定模型参数辨识的线性模型；If the identification result does not fit well with the frequency response curve, then return to step S2 to re-determine the linear model of the model parameter identification, and use the bipolar square wave in the VERIFY part to perform time domain verification on the linear model of the parameter identification of the model whose parameters have been determined. If the verification fails, it proves that the model obtained by the identification does not have the model prediction ability, and then returns to step S4 to re-determine the linear model of the model parameter identification; 重复步骤S6直至所有通道各个作动器操纵导数都被确定，得到完整的倾转旋翼无人机过渡模式模型。Step S6 is repeated until the manipulation derivatives of each actuator of all channels are determined, and a complete transition mode model of the tilt-rotor UAV is obtained.
CN202011016069.5A 2020-09-24 2020-09-24 A Tilting Rotor UAV Transition Mode Model Identification Method Expired - Fee Related CN112068582B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011016069.5A CN112068582B (en) 2020-09-24 2020-09-24 A Tilting Rotor UAV Transition Mode Model Identification Method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011016069.5A CN112068582B (en) 2020-09-24 2020-09-24 A Tilting Rotor UAV Transition Mode Model Identification Method

Publications (2)

Publication Number Publication Date
CN112068582A CN112068582A (en) 2020-12-11
CN112068582B true CN112068582B (en) 2022-02-18

Family

ID=73682775

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011016069.5A Expired - Fee Related CN112068582B (en) 2020-09-24 2020-09-24 A Tilting Rotor UAV Transition Mode Model Identification Method

Country Status (1)

Country Link
CN (1) CN112068582B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113189866B (en) * 2021-02-03 2022-04-26 北京航空航天大学 A state space model identification method for tilt-rotor UAV
CN115310247A (en) * 2021-05-06 2022-11-08 北京理工大学 Design and Frequency Domain Identification Method of UAV Hovering Experiment Platform
CN115309178B (en) * 2021-05-06 2024-11-29 北京理工大学 Design of a small quadrotor UAV forward flight experimental system and model identification method
CN113296533B (en) * 2021-05-21 2022-02-22 深圳市边界智控科技有限公司 Generalized actuator control allocation and reconfiguration method, device and related components thereof
CN114859952B (en) * 2022-05-07 2025-02-07 南京航空航天大学 A nonlinear incremental adaptive dynamic optimization control method for helicopters
CN114879510B (en) * 2022-05-31 2025-04-08 中山大学 A method and device for identifying a speed control loop model of an unmanned aerial vehicle

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103010485A (en) * 2012-12-21 2013-04-03 南京航空航天大学 Simulation modeling method for tilt-rotor unmanned plane and system thereof
CN107992070A (en) * 2017-12-03 2018-05-04 中国直升机设计研究所 A kind of tiltrotor aircraft transition mode Automatic implementation
CN111240212A (en) * 2020-03-25 2020-06-05 北京航空航天大学 A Tilting Rotor UAV Control Assignment Method Based on Optimal Prediction
CN111538237A (en) * 2020-03-20 2020-08-14 北京航空航天大学 Method for identifying and correcting non-linear light gray model of tilt rotor unmanned aerial vehicle

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8972310B2 (en) * 2012-03-12 2015-03-03 The Boeing Company Method for identifying structural deformation
CN103144781A (en) * 2012-12-21 2013-06-12 南京航空航天大学 Method for determining transient process switching corridor of tilt rotor unmanned aircraft
CN109946971B (en) * 2019-04-04 2021-09-24 南京航空航天大学 A smooth switching control method for the transition section of a tilt-rotor UAV

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103010485A (en) * 2012-12-21 2013-04-03 南京航空航天大学 Simulation modeling method for tilt-rotor unmanned plane and system thereof
CN107992070A (en) * 2017-12-03 2018-05-04 中国直升机设计研究所 A kind of tiltrotor aircraft transition mode Automatic implementation
CN111538237A (en) * 2020-03-20 2020-08-14 北京航空航天大学 Method for identifying and correcting non-linear light gray model of tilt rotor unmanned aerial vehicle
CN111240212A (en) * 2020-03-25 2020-06-05 北京航空航天大学 A Tilting Rotor UAV Control Assignment Method Based on Optimal Prediction

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
基于角加速度估计的非线性增量动态逆控制及试飞;党小为，等;《航空学报》;20200425;第41卷(第4期);第1-9页 *
小型农用无人直升机飞行控制技术研究;朱兖植;《中国优秀硕士学位论文全文数据库》;20140715(第7期);第C031-146页 *
旋翼/固定翼混合模式无人机在悬停状态下的模型辨识;孔维胜;《第29届中国控制与决策会议论文集（3）》;20170531;第1573-1579页 *
无人飞行器飞行控制仿真与软件设计;陈澄;《中国优秀硕士学位论文全文数据库》;20150115(第1期);第C031-139页 *

Also Published As

Publication number Publication date
CN112068582A (en) 2020-12-11

Similar Documents

Publication Publication Date Title
CN112068582B (en) A Tilting Rotor UAV Transition Mode Model Identification Method
CN111538237B (en) A nonlinear light gray model identification and correction method for tilt-rotor UAV
CN109270947A (en) Tilting rotor wing unmanned aerial vehicle flight control system
Silvestre et al. Aircraft control based on flexible aircraft dynamics
CN110334368A (en) A Flight Dynamics Modeling Method for Compound Thrust Configuration Helicopters
Su et al. Modeling and control of a class of urban air mobility tiltrotor aircraft
Takarics et al. Active flutter mitigation testing on the FLEXOP demonstrator aircraft
Juhasz et al. Flight dynamics simulation modeling of a large flexible tiltrotor aircraft
Simmons System identification for eVTOL aircraft using simulated flight data
CN112487551A (en) Multi-control-plane chain type direct control distribution and reconstruction method for flying wing unmanned aerial vehicle
Danowsky et al. Control-oriented system and parameter identification of a small flexible flying-wing aircraft
Hegde et al. Transition flight modeling and robust control of a VTOL unmanned quad tilt-rotor aerial vehicle
Saetti et al. Tiltrotor simulations with coupled flight dynamics, state-space aeromechanics, and aeroacoustics
Mystkowski 721. An application of mu-synthesis for control of a small air vehicle and simulation results
CN117634020A (en) Hypersonic aircraft overall control collaborative design method based on multi-fidelity data fusion
Kumar et al. Rotorcraft parameter identification from real time flight data
Berger et al. Advances and modern applications of frequency-domain aircraft and rotorcraft system identification
Vayalali et al. Horizontal stabilator utilization for post swashplate failure operation on a UH-60 black hawk helicopter
Tischler et al. Bell 412 system identification and model fidelity assessment for hover and forward flight
Hecker et al. Advanced gust load alleviation system for large flexible aircraft
Gilyard et al. Flight test of an adaptive configuration optimization system for transport aircraft
Nadell et al. System Identification and Stitched Modeling of the ADAPT™ Winged Compound Helicopter Scaled Demonstrator
CN113189866A (en) Method for identifying state space model of tilt rotor unmanned aerial vehicle
CN114545771B (en) A multi-mode adaptive switching control method and system for composite wing unmanned aerial vehicle
Altay et al. Experimental verification of a simulation model for jet UAV with model based design

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20220218