CN106200377A

CN106200377A - An Estimation Method of Aircraft Control Parameters

Info

Publication number: CN106200377A
Application number: CN201610497466.6A
Authority: CN
Inventors: 廖瑛; 郑宇昕; 文援兰; 何星星
Original assignee: National University of Defense Technology
Current assignee: National University of Defense Technology
Priority date: 2016-06-29
Filing date: 2016-06-29
Publication date: 2016-12-07

Abstract

The invention provides an estimation method of aircraft control parameters, which comprises the steps of firstly introducing a Constant adaptation model, establishing a state equation and a measurement equation of an aircraft control system, then analyzing the influence of a standard Kalman filtering algorithm rule and a system model error on a filtering result, adopting a method of dynamically adjusting filter estimation weight occupied by prediction information, giving a criterion and a recursion formula of an robust adaptive Kalman filtering algorithm, finally taking actual control parameter data as system input, and estimating a control number by using the robust adaptive Kalman filtering algorithm. The aircraft control parameters estimated by the method have better robustness and estimation accuracy than a standard Kalman filtering algorithm. And an effective means is provided for the engineering realization of aircraft parameter estimation.

Description

An Estimation Method of Aircraft Control Parameters

技术领域technical field

本发明涉及参数辨识技术领域，具体的涉及一种飞行器控制参数的估计方法。The invention relates to the technical field of parameter identification, in particular to a method for estimating control parameters of an aircraft.

背景技术Background technique

随着航空航天技术的飞速发展，对飞行器气动力性能指标提出了越来越高的要求，准确预测空气动力特性是设计高性能飞行器控制系统的基础和前提，而气动参数在飞行器运动过程中是随着运动环境和条件的变化而变化的，并且在实际飞行全过程中控制系统本身以及飞行条件的复杂性、传感器及数据采集的非理想性、飞行试验设计的不完善性等因素的影响，使得飞行器系统模型中的气动参数存在偏差，而这种偏差必然会体现在飞行器的控制参数中，因此在实践中需要对可测的控制参数进行正确的估计。With the rapid development of aerospace technology, higher and higher requirements are put forward for aircraft aerodynamic performance indicators. Accurate prediction of aerodynamic characteristics is the basis and premise of designing high-performance aircraft control systems, and aerodynamic parameters are important in the process of aircraft movement. It changes with the change of the motion environment and conditions, and is affected by factors such as the complexity of the control system itself and flight conditions, the non-ideality of sensors and data acquisition, and the imperfection of flight test design during the entire actual flight process. The aerodynamic parameters in the aircraft system model have deviations, and this deviation will inevitably be reflected in the control parameters of the aircraft. Therefore, it is necessary to correctly estimate the measurable control parameters in practice.

基于Bayes理论的Kalman滤波，需要获取准确的模型先验信息，只有当系统模型和随机模型的先验精度均满足条件时，滤波估计才会得到具有良好统计特性的最优解。但飞行器在处于动态运动过程中时，存在模型不确定及外界干扰，这使得先验信息的统计结果可能失真而不能够直接使用，即使使用了也会导致构建的状态模型与实际中的模型存在较大差异，这将影响到参数估计结果的质量。在已有的研究成果中，对飞行器控制参数在动态运动状态下的估计精度并不理想。The Kalman filter based on Bayesian theory needs to obtain accurate model prior information. Only when the prior accuracy of the system model and the random model meet the conditions, the filter estimation can obtain the optimal solution with good statistical characteristics. However, when the aircraft is in the process of dynamic motion, there are model uncertainties and external interference, which may distort the statistical results of prior information and cannot be used directly. Larger differences will affect the quality of parameter estimation results. In the existing research results, the estimation accuracy of aircraft control parameters in dynamic motion state is not ideal.

发明内容Contents of the invention

本发明的目的在于提供一种飞行器控制参数的估计方法，该发明解决了现有技术中飞行器控制参数在动态运动状态下的估计精度不高的技术问题。The object of the present invention is to provide a method for estimating aircraft control parameters, which solves the technical problem in the prior art that the estimation accuracy of aircraft control parameters is not high in a dynamic motion state.

本发明提供了一种飞行器控制参数的估计方法，估计过程如图1所示，包括以下步骤：The present invention provides a method for estimating aircraft control parameters, the estimation process as shown in Figure 1, comprising the following steps:

步骤S100：选择需处理的飞行器控制参数，并测量其实际值；Step S100: Select the aircraft control parameters to be processed, and measure their actual values;

步骤S200：根据动态系统的特点借鉴Constant Acceleration(CA)模型，建立飞行器控制系统中测量参数与时间的状态方程、观测方程；Step S200: learn from the Constant Acceleration (CA) model according to the characteristics of the dynamic system, and establish state equations and observation equations for measuring parameters and time in the aircraft control system;

步骤S300：构建抗差自适应Kalman滤波算法，在算法中增加自适应因子α_k调整其在参数估计中的贡献，同时增加抗差等价权矩阵抗差等价权矩阵是观测向量权矩阵的自适应估值。Step S300: Build a robust adaptive Kalman filter algorithm, add an adaptive factor α _k to the algorithm to adjust its contribution in parameter estimation, and increase the robust equivalent weight matrix at the same time Resilience Equivalent Weight Matrix is the adaptive estimate of the observation vector weight matrix.

步骤S400：根据步骤二中建立的飞行器控制系统，应用步骤三中所述的抗差自适应Kalman滤波算法对飞行器控制参数的测量值进行参数估计，得到其估计值。为了验证所得估计值的精度，计算飞行器控制参数估计值与其真实值的误差量，重复步骤S300和步骤S400直至所得估计值和实际值的误差满足精度要求时，停止迭代，输出所得估计值；。Step S400: According to the aircraft control system established in step 2, apply the robust adaptive Kalman filtering algorithm described in step 3 to estimate the measured values of the aircraft control parameters to obtain the estimated values. In order to verify the accuracy of the obtained estimated value, calculate the error amount between the estimated value of the aircraft control parameter and its true value, repeat steps S300 and S400 until the error between the obtained estimated value and the actual value meets the accuracy requirement, stop the iteration, and output the obtained estimated value;

其中，在步骤S200中所述的动态系统状态方程及观测方程的建立包括以下步骤：Wherein, the establishment of the dynamic system state equation and the observation equation described in step S200 includes the following steps:

将飞行器控制参数用x表示，并假设控制参数与时间t可用非线性函数表示如下：Express the control parameters of the aircraft with x, and assume that the control parameters and time t can be expressed as a nonlinear function as follows:

x＝x(t) (1)x=x(t) (1)

根据CA模型，在有限时间内，飞行器控制参数可用时间的2阶Taylor展开近似，设测量参数采样时间间隔为Δt，则其CA模型如下，并在此基础上建立飞行器控制系统的状态方程。According to the CA model, within a finite time, the second-order Taylor expansion approximation of the available time of aircraft control parameters, assuming that the measurement parameter sampling time interval is Δt, the CA model is as follows, and the state equation of the aircraft control system is established on this basis.

${x x}_{k k + + 11} = = {x x}_{k k} + + {\overset{\cdot &Center Dot;}{x x}}_{k k} Δ Δ t t + + {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \frac{{Δt Δt}^{22}}{22} + + O o (({Δt Δt}^{33})) - - - - - - ((22))$

其中k代表第k个采样时刻；Where k represents the kth sampling moment;

由(2)式可得飞行器控制系统的状态方程：The state equation of the aircraft control system can be obtained from formula (2):

$[\begin{matrix} {x x}_{k k} \\ {\overset{\cdot &Center Dot;}{x x}}_{k k} \\ {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}] = = [\begin{matrix} 11 & Δ Δ t t & {Δt Δt}^{22} / / 22 \\ 00 & 11 & Δ Δ t t \\ 00 & 00 & 11 \end{matrix}] [\begin{matrix} {x x}_{k k} \\ {\overset{\cdot \cdot}{x x}}_{k k} \\ {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}] + + {w w}_{k k} - - - - - - ((33))$

式中，状态量x_k表示飞行器控制系统控制参数，即是需要估计的参数，表示控制参数随时间的变化率，表示控制参数随时间变化的加速度；w_k＝[w_k1 w_k2 w_k3]^T代表各状态量的误差，即飞行器控制系统的模型噪声。In the formula, the state quantity x _k represents the control parameters of the aircraft control system, that is, the parameters that need to be estimated, Indicates the rate of change of the control parameter with time, Indicates the acceleration of control parameters changing with time; w _k =[w _k1 w _k2 w _k3 ] ^T represents the error of each state quantity, that is, the model noise of the aircraft control system.

飞行器控制系统的观测方程记为：The observation equation of the aircraft control system is recorded as:

${y the y}_{k k} = = {H h}_{k k} {[\begin{matrix} {x x}_{k k} & {\overset{\cdot &Center Dot;}{x x}}_{k k} & {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}]}^{T T} + + {v v}_{k k} - - - - - - ((44))$

式中，v_k表示测量误差，本文将采用抗差自适应Kalman滤波算法估计飞行器控制参数，并设计观测矩阵H_k为1×3维，即H_k＝[1 Δt Δt²/2]。In the formula, v _k represents the measurement error. In this paper, the robust adaptive Kalman filter algorithm is used to estimate the control parameters of the aircraft, and the observation matrix H _k is designed to be 1 × 3-dimensional, that is, H _k = [1 Δt Δt ² /2].

其中，步骤S300中构建抗差自适应Kalman滤波算法，包括以下步骤：Wherein, in the step S300, constructing a robust adaptive Kalman filter algorithm includes the following steps:

(1)抗差自适应Kalman滤波估计准则(1) Robust Adaptive Kalman Filter Estimation Criterion

由标准Kalman滤波准则出发，分析模型误差对参数估计的影响，从而可导出抗差自适应Kalman滤波估计准则。Starting from the standard Kalman filter criterion, the impact of model errors on parameter estimation is analyzed, and the robust adaptive Kalman filter estimation criterion can be derived.

通常定义不显含时间t的自治系统的动力学方程：It is common to define the dynamical equations for autonomous systems without explicit time t:

X_k＝Φ_k,k-1X_k-1+W_k (5)X _k ＝Φ _k,k-1 X _k-1 +W _k (5)

定义其测量方程为：Define its measurement equation as:

Y_k＝H_kX_k+V_k (6)Y _k ＝H _k X _k +V _k (6)

其中，X_k为状态向量，Φ_k,k-1为状态转移矩阵，Y_k为观测值向量，W_k为状态模型噪声矩阵，其协方差矩阵为V_k为测量噪声，其协方差矩阵为Σ_k，并且，状态噪声W_k和测量噪声V_k为无不相关的白噪声序列。Among them, X _k is the state vector, Φ _k,k-1 is the state transition matrix, Y _k is the observation vector, W _k is the state model noise matrix, and its covariance matrix is V _k is measurement noise, its covariance matrix is Σ _k , and state noise W _k and measurement noise V _k are uncorrelated white noise sequences.

标准Kalman滤波算法的精确实现要求可靠的动力学模型、观测先验信息以及合理的估计方法。由其递推公式可知，滤波增益本质上是决定了观测值Y_k和状态预测值在状态估计中的比例权重。因此当观测精度较高而模型精度较低时，若仍旧按照标准Kalman滤波进行滤波估计，其估计结果将偏离真值，滤波很可能出现发散的情况。可以通过对滤波增益值大小的调节，适当的匹配观测值和预测值的权重，当模型存在异常误差时增加状态预测值的协方差矩阵权重从而降低了预测信息对参数估计的权重，增加观测信息对其的权重，达到较好的参数估计效果。Accurate implementation of standard Kalman filtering algorithm requires reliable dynamic model, observation prior information and reasonable estimation method. It can be seen from its recursive formula that the filter gain essentially determines the observed value Y _k and the state predicted value Scale weight in state estimation. Therefore, when the observation accuracy is high and the model accuracy is low, if the filter estimation is still performed according to the standard Kalman filter, the estimation result will deviate from the true value, and the filtering is likely to diverge. By adjusting the size of the filter gain value, the weight of the observed value and the predicted value can be appropriately matched, and the predicted value of the state can be increased when the model has abnormal errors. The covariance matrix weights of Therefore, the weight of the prediction information on the parameter estimation is reduced, and the weight of the observation information is increased to achieve a better parameter estimation effect.

综上，将模型信息作为一个整体，引入自适应因子α_k调整其在参数估计中的贡献，同时引入抗差等价权矩阵它是观测向量权矩阵的自适应估值，从而构建抗差自适应Kalman滤波准则。To sum up, taking the model information as a whole, the adaptive factor α _k is introduced to adjust its contribution in parameter estimation, and the robust equivalent weight matrix is introduced at the same time It is an adaptive estimation of the weight matrix of the observation vector, so as to construct a robust adaptive Kalman filtering criterion.

设状态预测信息向量的误差方程为Let the error equation of the state prediction information vector be

${V V}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = {\overset{^^}{X x}}_{k k} - - {\overset{&OverBar; &OverBar;}{X x}}_{k k} = = {\overset{^^}{X x}}_{k k} - - {Φ Φ}_{k k,, k k - - 11} {\overset{&OverBar; &OverBar;}{X x}}_{k k} - - - - - - ((77))$

式中，为状态参数的估计值。则在t_k时刻，观测值估计残差向量为In the formula, is the estimated value of the state parameter. Then at time t _k , the estimated residual vector of observations is

${V V}_{k k} = = {H h}_{k k} {\overset{^^}{X x}}_{k k} - - {Y Y}_{k k} - - - - - - ((88))$

为控制观测异常和状态预测信息异常对状态参数估计值的影响，改造如下抗差自适应滤波原则：In order to control the influence of observation anomalies and state prediction information anomalies on state parameter estimates, the robust adaptive filtering principle is modified as follows:

$Ω Ω = = {V V}_{k k}^{T T} {\overset{&OverBar; &OverBar;}{P P}}_{k k} {V V}_{k k} + + {α α}_{k k} {V V}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}}^{T T} {P P}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {V V}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = m m i i n no - - - - - - ((99))$

式中，为Y_k的抗差等价权矩阵，是观测向量权矩阵的自适应估值；α_k为自适应因子，通常取0≤α_k≤1。In the formula, is the robust equivalent weight matrix of Y _k , which is the adaptive estimation of the weight matrix of the observation vector; α _k is the adaptive factor, usually 0≤α _k ≤1.

(2)抗差自适应Kalman滤波递推公式(2) Robust Adaptive Kalman Filter Recursion Formula

根据(9)式定义抗差自适应Kalman滤波准则，由条件极值原理，对(5)式和(6)式描述的自治系统，仿照标准Kalman滤波公式，可以得到抗差自适应Kalman滤波的递推公式：According to formula (9), the criterion of robust adaptive Kalman filter is defined, and the autonomous system described by formula (5) and formula (6) is modeled on the standard Kalman filter formula, and the robust adaptive Kalman filter can be obtained by the conditional extremum principle Recursion formula:

1)保存t_k-1时刻的及其协方差矩阵 1) Save time t _k-1 and its covariance matrix

2)状态预测：2) State prediction:

预测状态predictive status

${\overset{&OverBar; &OverBar;}{X x}}_{k k} = = {Φ Φ}_{k k,, k k - - 11} {\overset{^^}{X x}}_{k k - - 11} - - - - - - ((1010))$

预测状态协方差阵Predicted state covariance matrix

${Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = {Φ Φ}_{k k,, k k - - 11} {Σ Σ}_{{\overset{^^}{X x}}_{k k - - 11}} {Φ Φ}_{k k,, k k - - 11}^{T T} + + {Σ Σ}_{{W W}_{k k}} - - - - - - ((1111))$

3)计算观测值Y_k的抗差等价权矩阵 3) Calculate the robust equivalent weight matrix of the observed value Y _k

4)计算整体自适应因子α_k(0≤α_k≤1)4) Calculate the overall adaptive factor α _k (0≤α _k ≤1)

5)观测更新：5) Observation update:

新息向量innovation vector

${\overset{&OverBar; &OverBar;}{V V}}_{k k} = = {H h}_{k k} {\overset{&OverBar; &OverBar;}{X x}}_{k k} - - {Y Y}_{k k} - - - - - - ((1212))$

新息向量协方差矩阵Innovation vector covariance matrix

${Σ Σ}_{{\overset{&OverBar; &OverBar;}{V V}}_{k k}} = = {H h}_{k k} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} + + {Σ Σ}_{k k} - - - - - - ((1313))$

增益矩阵gain matrix

${\overset{&OverBar; &OverBar;}{K K}}_{k k} = = \frac{11}{{α α}_{k k}} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} {((\frac{11}{{α α}_{k k}} {H h}_{k k} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} + + {\overset{&OverBar; &OverBar;}{Σ Σ}}_{k k}))}^{- - 11} - - - - - - ((1414))$

状态估计向量state estimation vector

${\overset{^^}{X x}}_{k k} = = ((I I - - {\overset{&OverBar; &OverBar;}{K K}}_{k k} {H h}_{k k})) {\overset{&OverBar; &OverBar;}{X x}}_{k k} + + {\overset{&OverBar; &OverBar;}{K K}}_{k k} {Y Y}_{k k} - - - - - - ((1515))$

状态估计向量协方差矩阵State estimation vector covariance matrix

${Σ Σ}_{{\overset{^^}{X x}}_{k k}} = = ((I I - - {\overset{&OverBar; &OverBar;}{K K}}_{k k} {H h}_{k k})) {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} / / {α α}_{k k} - - - - - - ((1616))$

6)令k＝k+1，回到第1)步，重复上述过程，直到所得估计结果满足精度要求，算法停止。此处的迭代结束是指迭代计算结果满足精度要求时，此处的精度要求可以根据实际情况进行设定。例如可以设定迭代结束条件为迭代计算次数达到5000次时。6) Set k=k+1, return to step 1), repeat the above process until the obtained estimation result meets the accuracy requirement, and the algorithm stops. The iteration end here refers to when the iterative calculation result meets the precision requirement, and the precision requirement here can be set according to the actual situation. For example, the iteration end condition may be set as when the number of iteration calculations reaches 5000 times.

从上述抗差自适应滤波推导公式中可以看出，当观测信息含有异常，算法会自动将等价权矩阵减小，降低其权值，从而控制异常观测对参数估计的影响；当模型产生异常扰动时，相对应的α_k减小，从而控制状态预测信息对参数估计的影响。From the above derivation formula of robust adaptive filtering, it can be seen that when the observation information contains anomalies, the algorithm will automatically reduce the equivalent weight matrix and reduce its weight, thereby controlling the influence of abnormal observations on parameter estimation; when the model produces anomalies When disturbed, the corresponding α _k decreases, thereby controlling the influence of state prediction information on parameter estimation.

(3)抗差等价权矩阵(3) Resilience Equivalent Weight Matrix

实践中，可以直接构建抗差等价权函数，如Huber权函数、丹麦法权函数等，考虑测量的先验信息，采用IGGⅢ方案，其对应的权函数采用三段法。In practice, it is possible to directly construct the robust equivalent weight function, such as the Huber weight function, the Danish law weight function, etc., considering the prior information of the measurement, using the IGGⅢ scheme, and the corresponding weight function adopts the three-stage method.

${\overset{&OverBar; &OverBar;}{P P}}_{k k} ((i i)) = = \{\begin{matrix} {P P}_{k k} ((i i)) & | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {k k}_{00} \\ {P P}_{k k} ((i i)) \frac{{k k}_{00}}{| | {\overset{~ ~}{V V}}_{k k} | |} {((\frac{{k k}_{11} - - | | | | {\overset{~ ~}{V V}}_{k k} | | | |}{{k k}_{11} - - {k k}_{00}}))}^{22} & {k k}_{00} \leq \leq | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {k k}_{11} \\ 00 & | | {\overset{~ ~}{V V}}_{k k} | | &GreaterEqual; &Greater Equal; {k k}_{11} \end{matrix} - - - - - - ((1717))$

式中，k₀可取1.0～1.5，k₁可取2.5～8.0；为标准化预测残差：In the formula, k ₀ can be 1.0~1.5, k ₁ can be 2.5~8.0; For standardized prediction residuals:

${\overset{~ ~}{V V}}_{k k} | | | | {\overset{&OverBar; &OverBar;}{V V}}_{k k} | | | | / / \sqrt{t t r r a a c c e e (({Σ Σ}_{{\overset{&OverBar; &OverBar;}{V V}}_{k k}}))} - - - - - - ((1818))$

根据抗差估计理论，可以利用向量来判断模型是否存在显著异常。According to the robust estimation theory, the vector To judge whether there is a significant abnormality in the model.

(4)自适应因子(4) Adaptive factor

如前所述，在抗差自适应Kalman滤波中，可以通过自适应因子的构建，调节观测信息和预测信息对参数估计的贡献。对于单一自适应因子函数的构建，目前主要存在三段函数法、两段函数法及指数函数法，本发明中采用三段函数法进行构建。同时，由于预测残差向量能够较好的反应模型误差，这里采用基于预测残差的三段函数法构造自适应因子α_k：As mentioned above, in robust adaptive Kalman filtering, the contribution of observation information and prediction information to parameter estimation can be adjusted through the construction of adaptive factors. For the construction of a single self-adaptive factor function, currently there are mainly three-segment function method, two-segment function method and exponential function method, and the three-segment function method is used for construction in the present invention. At the same time, since the prediction residual vector To better reflect the model error, the adaptive factor α _k is constructed using the three-segment function method based on the prediction residual:

${α α}_{k k} = = \{\begin{matrix} 11 & | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {c c}_{00} \\ \frac{{c c}_{00}}{| | {\overset{~ ~}{V V}}_{k k} | |} {((\frac{{c c}_{11} - - | | | | {\overset{~ ~}{V V}}_{k k} | | | |}{{c c}_{11} - - {c c}_{00}}))}^{22} & {c c}_{00} \leq \leq | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {c c}_{11} \\ 00 & | | {\overset{~ ~}{V V}}_{k k} | | &GreaterEqual; &Greater Equal; {c c}_{11} \end{matrix} - - - - - - ((1919))$

式中，c₀取1.0～1.5，c₁取3.0～8.5。c₀和c₁的取值，也可以根据实际情况进行设置。In the formula, c ₀ is 1.0 to 1.5, and c ₁ is 3.0 to 8.5. The values of c ₀ and c ₁ may also be set according to actual conditions.

其中，在步骤S400中所述的飞行器控制参数估计中，选择飞行器控制参数作为待估计的变量，即步骤S300中的状态向量X_k，根据S300所提出的抗差自适应Kalman滤波估计方法，对飞行器控制参数进行估计，得到并计算估计值与实际值的误差。Wherein, in the aircraft control parameter estimation described in step S400, the aircraft control parameter is selected as the variable to be estimated, that is, the state vector X _k in step S300, according to the robust adaptive Kalman filter estimation method proposed in S300, for The control parameters of the aircraft are estimated to obtain And calculate the error between the estimated value and the actual value.

本发明的技术效果：Technical effect of the present invention:

1、本发明提供的飞行器控制参数的估计方法，以控制参数的状态模型和测量模型为估计对象，采用抗差自适应Kalman滤波算法，实现对控制模型不精确时的控制参数估计，通过引入自适应因子和抗差等价权矩阵，从而有效的提高了参数估计的精度。1. The method for estimating aircraft control parameters provided by the present invention takes the state model and measurement model of the control parameters as the estimation object, adopts the robust adaptive Kalman filter algorithm, and realizes the control parameter estimation when the control model is inaccurate. The adaptation factor and the robust equivalent weight matrix effectively improve the accuracy of parameter estimation.

2、本发明提供的飞行器控制参数的估计方法，能够准确辨识飞行器控制参数，具有良好的鲁棒性和估计精度，为飞行器控制参数估计的工程实现提供了有效手段。2. The method for estimating aircraft control parameters provided by the present invention can accurately identify aircraft control parameters, has good robustness and estimation accuracy, and provides an effective means for engineering realization of aircraft control parameter estimation.

3、本发明提供的飞行器控制参数的估计方法，通过三段函数法求解自适应因子，抗差自适应Kalman滤波可在LS估计(α_k＝0，)、标准Kalman滤波(α_k＝1，)、最小二乘自适应Kalman滤波(0≤α_k≤1，)、抗差Kalman滤波(α_k＝1)和抗差自适应滤波中自适应的选择最合适的滤波算法，因而具有较好的自适应效果。3. The method for estimating the aircraft control parameters provided by the present invention solves the adaptive factor by the three-section function method, and the robust adaptive Kalman filter can be estimated in LS (α _k =0, ), standard Kalman filtering (α _k =1, ), least square adaptive Kalman filter (0≤α _k ≤1, ), robust Kalman filter (α _k = 1) and robust adaptive filter to select the most suitable filtering algorithm adaptively, so it has better adaptive effect.

具体请参考根据本发明的飞行器控制参数的估计方法提出的各种实施例的如下描述，将使得本发明的上述和其他方面显而易见。For details, please refer to the following descriptions of various embodiments proposed according to the method for estimating aircraft control parameters of the present invention, so that the above and other aspects of the present invention will be apparent.

附图说明Description of drawings

图1是本发明提供的飞行器控制参数的估计方法流程示意图；Fig. 1 is a schematic flow chart of the method for estimating aircraft control parameters provided by the present invention;

图2是本发明优选实施例中所用某型空对地导弹控制模型中可测攻角模拟值曲线图；Fig. 2 is a curve diagram of measurable angle of attack analog values in a certain type of air-to-ground missile control model used in the preferred embodiment of the present invention;

图3是采用标准Kalman滤波对比例所得估计值与真实值的对比曲线图；Fig. 3 is the comparison curve diagram of estimated value and real value obtained by standard Kalman filtering comparative ratio;

图4是采用标准Kalman滤波对比例估计误差曲线图；Fig. 4 is a curve diagram of the error of the ratio estimation using the standard Kalman filter;

图5是本发明优选实施例中采用飞行器控制参数的估计方法所得估计值与真实值的对比曲线图；Fig. 5 is a comparison graph between the estimated value and the real value obtained by using the estimation method of the aircraft control parameter in the preferred embodiment of the present invention;

图6是本发明优选实施例中采用飞行器控制参数的估计方法所得估计误差曲线图。Fig. 6 is a curve diagram of estimation errors obtained by using an estimation method of aircraft control parameters in a preferred embodiment of the present invention.

具体实施方式detailed description

构成本申请的一部分的附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。The accompanying drawings constituting a part of this application are used to provide further understanding of the present invention, and the schematic embodiments and descriptions of the present invention are used to explain the present invention, and do not constitute an improper limitation of the present invention.

以下结合具体实施例对本发明所提供飞行器控制参数的估计方法进行详细说明。The method for estimating the aircraft control parameters provided by the present invention will be described in detail below in conjunction with specific embodiments.

步骤S100：以某型空对地导弹控制模型为例，由理想飞行状态计算得到攻角α随时间的变化值作为额定值，加入适当的偏离量，并根据传感器的噪声水平加入白噪声，从而模拟攻角的实际值。仿真某动态过程，其中偏离量为0.6°，设置采样步长为20ms，模拟产生10s的观测数据，攻角测量模拟值如图2所示。Step S100: Taking the control model of a certain type of air-to-ground missile as an example, the change value of the angle of attack α over time is calculated from the ideal flight state as the rated value, and an appropriate deviation is added, and white noise is added according to the noise level of the sensor, so that Simulates the actual value of the angle of attack. A dynamic process is simulated, where the deviation is 0.6°, the sampling step is set to 20ms, and the observation data of 10s is simulated. The simulated value of the angle of attack measurement is shown in Figure 2.

步骤S200：动态系统状态方程及观测方程的建立如下：Step S200: The dynamic system state equation and observation equation are established as follows:

飞行器控制系统的状态方程：The state equation of the aircraft control system:

$[\begin{matrix} {x x}_{k k} \\ {\overset{\cdot &Center Dot;}{x x}}_{k k} \\ {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}] = = [\begin{matrix} 11 & Δ Δ t t & {Δt Δt}^{22} / / 22 \\ 00 & 11 & Δ Δ t t \\ 00 & 00 & 11 \end{matrix}] [\begin{matrix} {x x}_{k k} \\ {\overset{\cdot &Center Dot;}{x x}}_{k k} \\ {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}] + + {w w}_{k k} - - - - - - ((2020))$

其中，w_k＝[w_k1 w_k2 w_k3]^T＝[0.01 0.01 0.01]^T，状态变量滤波初值设为0。Wherein, w _k ＝[w _k1 w _k2 w _k3 ] ^T ＝[0.01 0.01 0.01] ^T , and the initial value of state variable filtering is set to 0.

${y the y}_{k k} = = {H h}_{k k} {[\begin{matrix} {x x}_{k k} & {\overset{\cdot &Center Dot;}{x x}}_{k k} & {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}]}^{T T} + + {v v}_{k k} - - - - - - ((21 twenty one))$

式中，v_k＝1.2²，H_k＝[1 Δt Δt²/2]，Δt＝20ms。In the formula, v _k =1.2 ² , H _k =[1 Δt Δt ² /2], Δt=20ms.

步骤S300：抗差自适应Kalman滤波设计：引入自适应因子α_k调整其在参数估计中的贡献，同时引入抗差等价权矩阵从而构建抗差自适应Kalman滤波准则。抗差等价权矩阵的构造如下：Step S300: Robust adaptive Kalman filter design: introduce an adaptive factor α _k to adjust its contribution in parameter estimation, and introduce a robust equivalent weight matrix Therefore, the robust adaptive Kalman filter criterion is constructed. Resilience Equivalent Weight Matrix is constructed as follows:

${\overset{&OverBar; &OverBar;}{P P}}_{k k} ((i i)) = = \{\begin{matrix} {P P}_{k k} ((i i)) & | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {k k}_{00} \\ {P P}_{k k} ((i i)) \frac{{k k}_{00}}{| | {\overset{~ ~}{V V}}_{k k} | |} {((\frac{{k k}_{11} - - | | | | {\overset{~ ~}{V V}}_{k k} | | | |}{{k k}_{11} - - {k k}_{00}}))}^{22} & {k k}_{00} \leq \leq | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {k k}_{11} \\ 00 & | | {\overset{~ ~}{V V}}_{k k} | | &GreaterEqual; &Greater Equal; {k k}_{11} \end{matrix} - - - - - - ((22 twenty two))$

选取k₀＝1.5，k₁＝6.5。Choose k ₀ =1.5, k ₁ =6.5.

自适应因子α_k的构造如下：The adaptive factor α _k is constructed as follows:

${α α}_{k k} = = \{\begin{matrix} 11 & | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {c c}_{00} \\ \frac{{c c}_{00}}{| | {\overset{~ ~}{V V}}_{k k} | |} {((\frac{{c c}_{11} - - | | | | {\overset{~ ~}{V V}}_{k k} | | | |}{{c c}_{11} - - {c c}_{00}}))}^{22} & {c c}_{00} \leq \leq | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {c c}_{11} \\ 00 & | | {\overset{~ ~}{V V}}_{k k} | | &GreaterEqual; &Greater Equal; {c c}_{11} \end{matrix} - - - - - - ((23 twenty three))$

选取c₀＝1.5，c₁＝7.0。Choose c ₀ =1.5, c ₁ =7.0.

步骤S400：选择攻角α及其一阶导数、二阶导数为状态向量X_k，根据公式(10)～(19)计算攻角的估计值直到算法迭代到达5000次后停止。Step S400: Select the angle of attack α and its first-order derivative and second-order derivative as the state vector X _k , and calculate the estimated value of the attack angle according to formulas (10)-(19) Stop until the algorithm iteration reaches 5000 times.

为说明所得估计值的准确性计算误差量，计算指令攻角与估计攻角之间的误差量：e＝α_C-α。In order to illustrate the accuracy of the estimated value, the error amount is calculated, and the error amount between the commanded angle of attack and the estimated angle of attack is calculated: e = α _C - α.

为了与本发明的参数估计方法进行对比，以标准Kalman滤波算法为对比例在相同条件下对飞行器控制参数进行数值仿真，对比例中的相关参数设置与抗差自适应Kalman滤波相同。标准Kalman滤波算法详见“杨元喜.自适应动态导航定位[M].北京:测绘出版社,2006.”，具体步骤如下。In order to compare with the parameter estimation method of the present invention, the standard Kalman filter algorithm is used as a comparison example to carry out numerical simulation on the control parameters of the aircraft under the same conditions, and the relevant parameter settings in the comparison example are the same as the robust adaptive Kalman filter. For the standard Kalman filtering algorithm, see "Yang Yuanxi. Adaptive Dynamic Navigation and Positioning [M]. Beijing: Surveying and Mapping Press, 2006." The specific steps are as follows.

(1)保存t_k-1时刻的及其协方差矩阵 (1) Save time t _k-1 and its covariance matrix

(2)状态预测：(2) State prediction:

${\overset{&OverBar; &OverBar;}{X x}}_{k k} = = {Φ Φ}_{k k,, k k - - 11} {\overset{^^}{X x}}_{k k - - 11}$

${Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = {Φ Φ}_{k k,, k k - - 11} {Σ Σ}_{{\overset{^^}{X x}}_{k k - - 11}} {Φ Φ}_{k k,, k k - - 11}^{T T} + + {Σ Σ}_{{W W}_{k k}}$

(3)观测更新：(3) Observation update:

${\overset{&OverBar; &OverBar;}{V V}}_{k k} = = {H h}_{k k} {\overset{&OverBar; &OverBar;}{X x}}_{k k} - - {Y Y}_{k k}$

${Σ Σ}_{{\overset{&OverBar; &OverBar;}{V V}}_{k k}} = = {H h}_{k k} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} + + {Σ Σ}_{k k}$

${K K}_{k k} = = {P P}_{{\overset{^^}{X x}}_{k k}}^{- - 11} {AP AP}_{k k}$

${\overset{^^}{X x}}_{k k} = = {\overset{&OverBar; &OverBar;}{X x}}_{k k} - - {K K}_{k k} {\overset{&OverBar; &OverBar;}{V V}}_{k k}$

${Σ Σ}_{{\overset{^^}{X x}}_{k k}} = = ((I I - - {K K}_{k k} {A A}_{k k})) {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k - - 11}} ((I I - - {A A}_{k k}^{T T} {K K}_{k k}^{T T})) + + {K K}_{k k} {Σ Σ}_{k k} {K K}_{k k}^{T T}$

(4)令k＝k+1，回到第(1)步。(4) Let k=k+1, return to step (1).

本发明提供飞行器控制参数的估计方法算例中飞行器控制参数估计结果如图5～6所示。图3～4为采用标准Kalman滤波算法对飞行器控制参数进行数值仿真的对比例估计结果，由图5～6可见：本发明所应用的参数估计方法能够较好的满足控制参数估计精度，说明了本发明提供方法的有效性。通过对比图3～4和图5～6可知：本发明提供的方法在飞行器动态运动或模型存在误差时，对控制参数的估计误差小、精度高，有效提高了参数估计的准确度。本发明提供方法采用抗差自适应Kalman滤波。The method for estimating the control parameters of the aircraft provided by the present invention results in the estimation of the control parameters of the aircraft in the examples shown in Figures 5-6. Fig. 3～4 is the comparative estimation result that adopts standard Kalman filter algorithm to carry out numerical simulation to aircraft control parameter, can be seen by Fig. 5～6: the parameter estimation method that the present invention applies can satisfy control parameter estimation precision preferably, has illustrated The present invention provides the effectiveness of the method. By comparing Figs. 3-4 with Figs. 5-6, it can be known that the method provided by the present invention can estimate the control parameters with small errors and high precision when there are errors in the dynamic motion of the aircraft or in the model, effectively improving the accuracy of parameter estimation. The method provided by the invention adopts robust adaptive Kalman filtering.

本领域技术人员将清楚本发明的范围不限制于以上讨论的示例，有可能对其进行若干改变和修改，而不脱离所附权利要求书限定的本发明的范围。尽管己经在附图和说明书中详细图示和描述了本发明，但这样的说明和描述仅是说明或示意性的，而非限制性的。本发明并不限于所公开的实施例。It will be clear to a person skilled in the art that the scope of the present invention is not limited to the examples discussed above, but that several changes and modifications are possible without departing from the scope of the invention as defined in the appended claims. While the invention has been illustrated and described in detail in the drawings and description, such illustration and description are illustrative or exemplary only and not restrictive. The invention is not limited to the disclosed embodiments.

通过对附图，说明书和权利要求书的研究，在实施本发明时本领域技术人员可以理解和实现所公开的实施例的变形。在权利要求书中，术语“包括”不排除其他步骤或元素，而不定冠词“一个”或“一种”不排除多个。在彼此不同的从属权利要求中引用的某些措施的事实不意味着这些措施的组合不能被有利地使用。权利要求书中的任何参考标记不构成对本发明的范围的限制。Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the invention, from a study of the drawings, the specification and the claims. In the claims, the term "comprising" does not exclude other steps or elements, while the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims shall not be construed as limiting the scope of the invention.

Claims

1. an estimation method of aircraft control parameters, is characterized in that, comprises the following steps:

Step S100: measuring the actual value of the aircraft control command to be processed;

Step S200: Establish state equations and observation equations for measuring parameters and time in the aircraft control system;

Step S300: Constructing an equivalent weight matrix containing adaptive factor α _k and robustness Robust Adaptive Kalman Filtering Algorithm;

Step S400: According to the aircraft control system established in step S200, apply the robust adaptive Kalman filtering algorithm described in step S300 to perform parameter estimation on the measured value of the aircraft control parameters, obtain its estimated value, and calculate the estimated value and the actual value , repeat steps S300 and S400 until the error between the obtained estimated value and the actual value meets the accuracy requirement, stop the iteration, and output the obtained estimated value;

In the step S300, constructing a robust adaptive Kalman filter algorithm includes the following steps:

(1) Robust Adaptive Kalman Filter Estimation Criterion

Define the dynamical equations for an autonomous system without explicit time t:

X _k ＝Φ _k,k-1 X _k-1 +W _k (5)

Define its measurement equation as:

Y _k ＝H _k X _k +V _k (6)

Among them, X _k is the state vector, Φ _k,k-1 is the state transition matrix, Y _k is the observation vector, W _k is the state model noise matrix, and its covariance matrix is V _k is measurement noise, and its covariance matrix is Σ _k , where state noise W _k and measurement noise V _k are white noise sequences;

From the difference between the estimated value of the state vector and the state prediction value, the error equation of the state prediction information vector is obtained as:

{V V}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = {\overset{^^}{X x}}_{k k} - - {\overset{&OverBar; &OverBar;}{X x}}_{k k} = = {\overset{^^}{X x}}_{k k} - - {Φ Φ}_{k k,, k k - - 11} {\overset{&OverBar; &OverBar;}{X x}}_{k k} - - - - - - ((77))

In the formula, is the estimated value of the state parameter;

At time t _k , the estimated value of the state vector is brought into formula (6), and the estimated residual vector of the observed value can be obtained as:

{V V}_{k k} = = {H h}_{k k} {\overset{^^}{X x}}_{k k} - - {Y Y}_{k k} - - - - - - ((88))

Set the adaptive factor α _k in formula (8) to obtain the principle of robust adaptive filtering:

Ω Ω = = {V V}_{k k}^{T T} {\overset{&OverBar; &OverBar;}{P P}}_{k k} {V V}_{k k} + + {α α}_{k k} {V V}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}}^{T T} {P P}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {V V}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = m m i i n no - - - - - - ((99))

in, is the robust equivalent weight matrix of Y _k , which is the adaptive estimation of the weight matrix of the observation vector; α _k is the adaptive factor, usually 0≤α _k ≤1;

(2) Robust Adaptive Kalman Filter Recursion Formula

Define the robust adaptive Kalman filter criterion according to formula (9), and obtain the recursive formula of the robust adaptive Kalman filter, including the following steps:

1) Obtain the time of t _k-1 and its covariance matrix

2) State prediction:

predictive status

{\overset{&OverBar; &OverBar;}{X x}}_{k k} = = {Φ Φ}_{k k,, k k - - 11} {\overset{^^}{X x}}_{k k - - 11} - - - - - - ((1010))

Predicted state covariance matrix

{Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} = = {Φ Φ}_{k k,, k k - - 11} {Σ Σ}_{{\overset{^^}{X x}}_{k k - - 11}} {Φ Φ}_{k k,, k k - - 11}^{T T} + + {Σ Σ}_{{W W}_{k k}} - - - - - - ((1111))

3) Calculate the robust equivalent weight matrix of the observed value Y _k

4) Calculate the overall adaptive factor α _k (0≤α _k ≤1)

5) Observation update:

innovation vector

{\overset{&OverBar; &OverBar;}{V V}}_{k k} = = {H h}_{k k} {\overset{&OverBar; &OverBar;}{X x}}_{k k} - - {Y Y}_{k k} - - - - - - ((1212))

Innovation vector covariance matrix

{Σ Σ}_{{\overset{&OverBar; &OverBar;}{V V}}_{k k}} = = {H h}_{k k} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} + + {Σ Σ}_{k k} - - - - - - ((1313))

gain matrix

{\overset{&OverBar; &OverBar;}{K K}}_{k k} = = \frac{11}{{α α}_{k k}} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} {((\frac{11}{{α α}_{k k}} {H h}_{k k} {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} {H h}_{k k}^{T T} + + {\overset{&OverBar; &OverBar;}{Σ Σ}}_{k k}))}^{- - 11} - - - - - - ((1414))

state estimation vector

{\overset{^^}{X x}}_{k k} = = ((I I - - {\overset{&OverBar; &OverBar;}{K K}}_{k k} {H h}_{k k})) {\overset{&OverBar; &OverBar;}{X x}}_{k k} + + {\overset{&OverBar; &OverBar;}{K K}}_{k k} {Y Y}_{k k} - - - - - - ((1515))

State estimation vector covariance matrix

{Σ Σ}_{{\overset{^^}{X x}}_{k k}} = = ((I I - - {\overset{&OverBar; &OverBar;}{K K}}_{k k} {H h}_{k k})) {Σ Σ}_{{\overset{&OverBar; &OverBar;}{X x}}_{k k}} / / {α α}_{k k} - - - - - - ((1616))

6) Make k=k+1, get back to step 1), repeat the above-mentioned process, and stop when the iteration ends;

(3) Using the IGGⅢ scheme to obtain the equivalent weight matrix of robustness

{\overset{&OverBar; &OverBar;}{P P}}_{k k} ((i i)) = = \{\begin{matrix} {P P}_{k k} ((i i)) & | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {k k}_{00} \\ {P P}_{k k} ((i i)) \frac{{k k}_{00}}{| | {\overset{~ ~}{V V}}_{k k} | |} {((\frac{{k k}_{11} - - | | | | {\overset{~ ~}{V V}}_{k k} | | | |}{{k k}_{11} - - {k k}_{00}}))}^{22} & {k k}_{00} \leq \leq | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {k k}_{11} \\ 00 & | | {\overset{~ ~}{V V}}_{k k} | | &GreaterEqual; &Greater Equal; {k k}_{11} \end{matrix} - - - - - - ((1717))

Among them, k ₀ is 1.0~1.5, k ₁ is 2.5~8.0; For standardized prediction residuals:

{\overset{~ ~}{V V}}_{k k} | | | | {\overset{&OverBar; &OverBar;}{V V}}_{k k} | | | | / / \sqrt{t t r r a a c c e e (({Σ Σ}_{{\overset{&OverBar; &OverBar;}{V V}}_{k k}}))} - - - - - - ((1818))

(4) The adaptive factor α _k is obtained by constructing the three-section function method based on the prediction residual:

{α α}_{k k} = = \{\begin{matrix} 11 & | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {c c}_{00} \\ \frac{{c c}_{00}}{| | {\overset{~ ~}{V V}}_{k k} | |} {((\frac{{c c}_{11} - - | | | | {\overset{~ ~}{V V}}_{k k} | | | |}{{c c}_{11} - - {c c}_{00}}))}^{22} & {c c}_{00} \leq \leq | | {\overset{~ ~}{V V}}_{k k} | | \leq \leq {c c}_{11} \\ 00 & | | {\overset{~ ~}{V V}}_{k k} | | &GreaterEqual; &Greater Equal; {c c}_{11} \end{matrix} - - - - - - ((1919))

Among them, c ₀ takes 1.0 to 1.5, and c ₁ takes 3.0 to 8.5.

2. The method for estimating aircraft control parameters according to claim 1, characterized in that, establishing state equations and observation equations of the aircraft control system in the step S200 comprises the following steps:

The relationship between aircraft control parameter x and time t satisfies:

x=x(t) (1)

In a finite time, the second-order Taylor expansion approximation of the available time of the aircraft control parameters is set, and the measurement parameter sampling time interval is Δt, and the CA model of the aircraft control parameter x is obtained:

{x x}_{k k + + 11} = = {x x}_{k k} + + {\overset{\cdot &Center Dot;}{x x}}_{k k} Δ Δ t t + + {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \frac{{Δt Δt}^{22}}{22} + + O o (({Δt Δt}^{33})) - - - - - - ((22))

Among them, k represents the kth sampling moment;

The state equation of the aircraft control system can be obtained from formula (2):

[\begin{matrix} {x x}_{k k} \\ {\overset{\cdot \cdot}{x x}}_{k k} \\ {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}] = = [\begin{matrix} 11 & Δ Δ t t & {Δt Δt}^{22} / / 22 \\ 00 & 11 & Δ Δ t t \\ 00 & 00 & 11 \end{matrix}] [\begin{matrix} {x x}_{k k} \\ {\overset{\cdot &Center Dot;}{x x}}_{k k} \\ {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}] + + {w w}_{k k} - - - - - - ((33))

Among them, the state quantity x _k represents the control parameters of the aircraft control system, Indicates the rate of change of the control parameter with time, Represents the acceleration of the control parameter changing with time; w _k =[w _k1 w _k2 w _k3 ] ^T represents the error of each state quantity, that is, the model noise of the aircraft control system;

According to the basic theory of Kalman filtering, the observation equation is recorded as:

{y the y}_{k k} = = {H h}_{k k} {[\begin{matrix} {x x}_{k k} & {\overset{\cdot \cdot}{x x}}_{k k} & {\overset{\cdot\cdot \cdot\cdot}{x x}}_{k k} \end{matrix}]}^{T T} + + {v v}_{k k} - - - - - - ((44))

Wherein, v _k represents a measurement error, and H _k is a 1×3-dimensional observation matrix, that is, H _k =[1 Δt Δt ² /2].

3. the estimation method of aircraft control parameter according to claim 2 is characterized in that, comprises the following steps in the described step S100: the angle of attack obtained by ideal flight state calculation changes with time as rated value, adds deviation amount , and white noise is added according to the noise level of the sensor used as the angle-of-attack measurement to simulate the actual angle-of-attack.