CN103708045B - The on-line parameter discrimination method that a kind of lunar exploration airship great-jump-forward reenters - Google Patents

Abstract

The invention discloses the on-line parameter discrimination method that a kind of lunar exploration airship great-jump-forward reenters, first descent stage is once being reentered and to be obtained by low-pass filter the On-line Estimation value of atmospheric density factor of proportionality and 1ift-drag ratio factor of proportionality, implement track forecast and revise offset landings, set up atmospheric density factor of proportionality and fitting function relation highly simultaneously; Once reentering ascent stage and secondary reentry phase initial stage, obtaining atmospheric density factor of proportionality by the fitting function set up, simultaneously the 1ift-drag ratio factor of proportionality of On-line Estimation lunar exploration airship, implementing track and forecast and revise offset landings; In the secondary reentry phase later stage, On-line Estimation atmospheric density factor of proportionality and 1ift-drag ratio factor of proportionality, implement track and forecast and revise offset landings.Invention increases robustness and the precision of guidance algorithm; Have the better comformability of Real Atmosphere model.

Description

The on-line parameter discrimination method that a kind of lunar exploration airship great-jump-forward reenters

Technical field

The present invention relates to Guidance and control technical field, can be applicable to the guidance that lunar exploration airship great-jump-forward reenters, particularly relate to the on-line parameter discrimination method that a kind of lunar exploration airship great-jump-forward reenters.

Background technology

Lunar exploration airship returns and reenters terrestrial time, the initial condition and the near-earth spacecraft that arrive air boundary have returned a great difference, the most obvious feature is that the reentry velocity of lunar exploration airship is larger, reach 11km/s, and the reentry velocity of near-earth spacecraft is generally about 7.8km/s.This high speed reentry condition make reentry trajectory to air and pneumatic parameter error more responsive, the Dynamics Coupling reentered in process is more serious, the dynamic pressure that airship is subject to, hot-fluid and overload change are more violent, and this all requires that reentry guidance algorithm has stronger robustness.Research shows: adopt great-jump-forward reentry mode can reduce overload in the process of reentering and hot-fluid peak value, the design pressure of reduction spacecraft structure and Thermal Protection System; Meanwhile, can increase and reenter voyage, improve the alerting ability that landing site is selected.Therefore, generally select great-jump-forward reentry mode when lunar exploration airship reenters, the Apollo Personnel investigation Program as the U.S. " airship, member's explorer vehicle (CEV) etc.

Reenter the medium-altitude change of process according to great-jump-forward, the whole process that reenters can be divided into three sections: reentry phase, exoatmosphere Kepler's section and a secondary reentry phase (accompanying drawing 2).In extraatmospheric Kepler's section, airship cannot control voyage by aerodynamic force, the voyage control ability of secondary reentry phase is also very limited, and the precision controlling key therefore reentered is a reentry phase, that is ensures that the end precision of a reentry phase is the core of guidance algorithm.Lunar exploration airship is subject to the impact of many error components when passing through earth atmosphere, comprise initial condition error, atmospheric density error, aerodynamic parameter error, quality error, navigation error, execution error etc., wherein aerodynamic parameter and atmospheric density two errors are major influence factors.Therefore, need the diffusion strictly suppressing this two classes error in guidance process, and compensated in Guidance Law by related algorithm.The present invention considers to adopt numerical prediction-correction method of guidance to return to ground smoothly to guide lunar exploration airship.

Although numerical prediction-correction guidance algorithm has stronger robustness, be difficult to the precision reaching expectation equally when error exceedes certain limit, aerodynamic parameter and atmospheric density error larger time particularly evident.Therefore, adopt error identification algorithm identification actual error, and identification result is fed back in guidance algorithm very necessary.Research shows, the error of aerodynamic parameter depends on the accuracy of ground wind tunnel test and numerical modelling, reenters this error change amplitude in process little, especially can be approximately constant value deviation under High Mach number.Atmospheric density error is then with solar activity, season, latitude, the Parameters variation such as round the clock, and the effect of experience Atmospheric Density Models when applying obtained by different observation method and theoretical analysis is unsatisfactory.Generally, atmospheric density error can regard the function of height H as, i.e. K _ρ=f (H), K _ρfor real atmosphere density and the ratio of standard atmosphere density.Therefore, under the prerequisite additionally not increasing sensor, utilize existing boat-carrying measuring equipment to come identification aerodynamic parameter and atmospheric density deviation, and feed back in guidance algorithm, thus the robustness of increase algorithm and precision have important engineering significance.。

Summary of the invention

Technical matters to be solved by this invention is, not enough for prior art, the on-line parameter discrimination method that a kind of lunar exploration airship great-jump-forward reenters is provided, overcome the interference of lunar exploration airship Aerodynamic Coefficient and atmospheric density deviation, improve robustness and the precision of numerical prediction-correction reentry guidance method, adapt to Real Atmosphere model better, thus instruct lunar exploration airship to fly better.

For solving the problems of the technologies described above, the technical solution adopted in the present invention is: the on-line parameter discrimination method that a kind of lunar exploration airship great-jump-forward reenters, and the method is:

1) when lunar exploration airship height drops to H _maxtime, start reentry guidance, the initial estimate of setting atmospheric density factor of proportionality with the initial estimate of airship 1ift-drag ratio

2) from first guidance period start time t of reentry guidance _1,1to once reentering a moment t that jumps up _{n, 1}time period in, any one guidance period start time t in this time period _{k, 1}, perform following steps:

2a) set current time as t _{k, 1}, measure and obtain t _{k, 1}the apparent acceleration of moment lunar exploration airship lunar exploration airship t is obtained according to inertial navigation principle _{k, 1}the flying height H in moment _{k, 1}and rate of change in altitude according to calculate t _{k, 1}the observed reading D of moment lunar exploration airship drag acceleration _{m_k, 1}with the observed reading L of lift acceleration/accel _{m_k, 1}, and then obtain t _{k, 1}the observed reading of moment 1ift-drag ratio ${\left(\frac{L}{D}\right)}_{m\_k,1}=\frac{L_{m\_k,1}}{D_{m\_k,1}};$

2b) by the observed reading D of drag acceleration _{m_k, 1}calculate t _{k, 1}the estimated valve ρ of moment atmospheric density _{m_k, 1}, according to lunar exploration airship t _{k, 1}the level height H in moment _{k, 1}obtain t _{k, 1}the atmospheric density ρ of moment nominal Atmospheric models _{0_k, 1}, thus obtain lunar exploration airship t _{k, 1}the atmospheric density factor of proportionality in moment $K_{\mathrm{\ρ}m\_k,1}=\frac{{\mathrm{\ρ}}_{m\_k,1}}{{\mathrm{\ρ}}_{0\_k,1}};$

The high frequency noise of 2c) filtering atmospheric density factor of proportionality and airship 1ift-drag ratio, obtains t _{k, 1}the estimated valve of moment atmospheric density factor of proportionality with the estimated valve of lunar exploration airship 1ift-drag ratio ${\left(\frac{L}{D}\right)}_{est}^{\[k,1\]}:$

$K_{\mathrm{\ρ}\_est}^{\[k,1\]}=(1-K_{g\mathrm{\ρ}})K_{\mathrm{\ρ}m\_k,1}+K_{g\mathrm{\ρ}}{K}_{\mathrm{\ρ}\_est}^{\[k-1,1\]};$

${\left(\frac{L}{D}\right)}_{est}^{\[k,1\]}=(1-K_{gLD}){\left(\frac{L}{D}\right)}_{m\_k,1}+K_{gLD}{\left(\frac{L}{D}\right)}_{est}^{\[k-1,1\]};$

In above formula, for t _{k, 1}a upper moment t in moment _k-1,1the estimated valve of moment atmospheric density factor of proportionality, initial value be k _{g ρ}for atmospheric density filtering factor, K _{ρ m_k, 1}for t _{k, 1}moment lunar exploration airship atmospheric density factor of proportionality; for t _k-1,1the estimated valve of moment lunar exploration airship 1ift-drag ratio, initial value be k _gLDfor 1ift-drag ratio filtering factor, for t _{k, 1}moment 1ift-drag ratio observed reading;

Then t _{k, 1}the 1ift-drag ratio factor of proportionality in moment computing formula be:

$K_{LD\_est}^{\[k,1\]}={\left(\frac{L}{D}\right)}_{est}^{\[k,1\]}/{\left(\frac{L}{D}\right)}_0^{\[k,1\]}$

In above formula, for t _{k, 1}the nominal 1ift-drag ratio of the lunar exploration airship that the moment is corresponding;

2d) store t _{k, 1}the flying height H of moment lunar exploration airship _{k, 1}with the estimated valve of atmospheric density factor of proportionality

2e) utilize numerical prediction-correction method of guidance to guide, forecasting process supposition atmospheric density factor of proportionality and 1ift-drag ratio factor of proportionality are constant value, obtain reentering lift acceleration/accel L in process _prewith drag acceleration D _preestimated valve; By dynam integration, forecast the flight path of lunar exploration airship, and then revise offset landings; In a forecasting process, from current time t _{k, 1}to the drag acceleration D reentering lunar exploration airship in end time _{pre_k+n}computing formula be:

$D_{pre\_k+n}=K_{\mathrm{\ρ}\_est}^{\[k,1\]}\frac{C_{D0\_k+n}{\mathrm{\ρ}}_{0\_k+n}{v}_{k+n}^{2}S}{2m};$

In above formula, m is the quality of lunar exploration airship; S is lunar exploration airship reference area; Subscript k+n represents from current time t _{k, 1}the n-th guidance cycle after rising, ρ _{0_k+n}for the atmospheric density obtained by ARDC model atmosphere ARDC, C _{d0_k+n}for the pneumatic drag coefficient obtained by aerodynamic data, v _k+nfor the speed of lunar exploration airship; In a forecasting process, from current t _{k, 1}moment is to the lift acceleration/accel L reentering lunar exploration airship in end time _{pre_k+n}computing formula be:

$L_{pre\_k+n}=K_{LD\_est}^{\[k,1\]}{K}_{\mathrm{\ρ}\_est}^{\[k,1\]}\frac{C_{L0\_k+n}{\mathrm{\ρ}}_{0\_k+n}{v}_{k+n}^{2}S}{2m};$

Wherein, C _{l0_k+n}for the aerodynamic lift coefficient obtained by aerodynamic data; C _{d0_k+n}, C _{l0_k+n}, ρ _{0_k+n}to be respectively in a forecasting process lunar exploration airship from current time t _{k, 1}to reentering the drag coefficient of end time, lift coefficient and atmospheric density;

2f) repeat above-mentioned steps 2a) ~ 2e), until the rate of change in altitude of lunar exploration airship be 0, namely lunar exploration airship arrive reentry trajectory jump up a little, enter step 3);

3) lunar exploration airship is recorded in described height H of jumping up a little _min, according to above-mentioned steps 2e) in from first of reentry guidance guidance period start time t _1,1to once reentering a moment t that jumps up _{n, 1}time period in store the altitude information H of lunar exploration airship and the estimated valve data K of atmospheric density factor of proportionality _{ρ _ est}, utilize the estimated valve K of following formula matching height H and atmospheric density factor of proportionality _{ρ _ est}between relation:

K _{ρ_est}＝A·H；

Wherein,

$H=\left[\begin{array}{cccc}1& 1& \mathrm{...}& 1\\ H_{1,1}& H_{2,1}& & H_{N,1}\\ H_{1,1}^2& H_{2,1}^2& & H_{N,1}^2\\ H_{1,1}^3& H_{2,1}^3& \mathrm{...}& H_{N,1}^3\\ H_{1,1}^4& H_{2,1}^4& & H_{N,1}^4\\ H_{1,1}^5& H_{2,1}^5& \mathrm{...}& H_{N,1}^5\end{array}\right];$

$K_{\mathrm{\ρ}\_est}=\left[\begin{array}{cccc}{K}_{\mathrm{\ρ}\_est}^{\[1,1\]},& K_{\mathrm{\ρ}\_est}^{\[2,1\]},& \mathrm{...},& K_{\mathrm{\ρ}\_est}^{\[N,1\]}\end{array}\right];$

A=[a ₀, a ₁, a ₂, a ₃, a ₄, a ₅], a ₀~ a ₅for fitting coefficient, can solve according to method of least square

A＝K _{ρ_est}·H ^T(HH ^T) ^-1

4) setting the initial estimate once reentering ascent stage lunar exploration airship 1ift-drag ratio is

5) once the ascent stage is reentered at lunar exploration airship, by first guidance period start time t _1,2h is arrived to lunar exploration airship flying height _maxmoment t _{n, 2}time period in, any instant t in this time period _{k, 2}in, perform following steps:

5a) measure the apparent acceleration W obtaining lunar exploration airship _{m_k, 2}, according to W _{m_k, 2}calculate the observed reading D of lunar exploration airship drag acceleration _{m_k, 2}with the observed reading L of lift acceleration/accel _{m_k, 2}, and then obtain the observed reading of 1ift-drag ratio the high frequency noise of the observed reading of filtering 1ift-drag ratio, obtains the estimated valve of lunar exploration airship 1ift-drag ratio

${\left(\frac{L}{D}\right)}_{est}^{\[k,2\]}=\left(1-K_{gLD}\right){\left(\frac{L}{D}\right)}_{m\_k,2}+K_{gLD}{\left(\frac{L}{D}\right)}_{est}^{\[k-1,2\]}$

In above formula, for t _{k, 2}a upper moment t in moment _k-1,2moment lunar exploration airship 1ift-drag ratio estimated valve, initial value be k _gLDfor 1ift-drag ratio filtering factor, for t _{k, 2}moment lunar exploration airship 1ift-drag ratio observed reading; Then t _{k, 2}moment 1ift-drag ratio factor of proportionality estimated valve for t _{k, 2}the nominal 1ift-drag ratio of the lunar exploration airship that the moment is corresponding;

5b) following formula is utilized to solve t _{k, 2}moment atmospheric density factor of proportionality estimated valve

$K_{\mathrm{\ρ}\_est}^{\[k,2\]}=a_0+a_1{H}_{k,2}+a_2{H}_{k,2}^2+a_3{H}_{k,2}^3+a_4{H}_{k,2}^4+a_5{H}_{k,2}^5;$

H _{k, 2}for t _{k, 2}the flying height of moment lunar exploration airship; Work as H _k>H _maxtime, ignore atmospheric influence; Work as H _{k, 2}<H _mintime, for lunar exploration airship is H once reentering a height of jumping up _minthe estimated valve of the atmospheric density factor of proportionality at place;

5c) utilize numerical prediction-correction method of guidance, assuming that 1ift-drag ratio factor of proportionality is constant value, from current time to reentering the drag acceleration before end in a forecasting process under utilizing following formula to estimate disturbance situation

$D_{pre\_H_k}=K_{\mathrm{\&rho;}\_est}^{H_k}\frac{C_{D0\_H_k}{\mathrm{\&rho;}}_{0\_H_k}{v}_{H_k}^{2}S}{2m};$

for present level H _kunder pneumatic drag coefficient, can be obtained by aerodynamic data table interpolation that airship stores; for present level H _kthe atmospheric density at place, can calculate according to Atmospheric models; for lunar exploration airship speed, can be provided by navigationsystem; for present level H _kunder the estimated valve of atmospheric density factor of proportionality;

From current time to reentering the lift acceleration/accel before end in a forecasting process under utilizing following formula to estimate disturbance situation

$L_{pre\_H_k}=K_{LD\_est}^{\&lsqb;k,2\&rsqb;}*K_{\mathrm{\&rho;}\_est}^{H_k}\frac{C_{L0\_H_k}{\mathrm{\&rho;}}_{0\_H_k}{v}_{H_k}^{2}S}{2m};$

for present level H _kthe aerodynamic lift coefficient that place is corresponding, can be obtained by the aerodynamic data table interpolation that airship stores;

The lunar exploration airship drag acceleration obtained will be estimated with lift acceleration/accel substitute in kinetics equation and carry out integration, obtain the lunar exploration airship flight path predicted, and guidance command reduction offset landings by what revise current time;

6) jump up when lunar exploration airship and be highly greater than H _maxtime, guidance system quits work; Arrive the vertex of the section of jumping up when lunar exploration airship and again drop to H _maxafter, secondary reentry guidance section starts, and at the secondary reentry guidance section initial stage, namely lunar exploration airship height is greater than H _mintime, guidance system is according to step 5) works; In the secondary reentry guidance section later stage, namely lunar exploration airship height is less than H _mintime, guidance system is according to step 2) works, until lunar exploration airship arrives parachute-opening point H _end.

Compared with prior art, the beneficial effect that the present invention has is: on-line identification atmospheric density deviation of the present invention and airship Aerodynamic Coefficient deviation, be applied to identification result in numerical prediction-correction method of guidance, improves robustness and the precision of guidance algorithm; The present invention utilizes the data once reentering descent stage to carry out fitting of a polynomial to atmospheric density factor of proportionality, the coefficient of polynomial fitting is solved by least-squares algorithm, once reenter ascent stage and secondary reentry phase just section utilize model of fit to implement to forecast at line tracking, compare the forecasting procedure adopting constant value atmospheric density factor of proportionality, have the better comformability of Real Atmosphere model.

Accompanying drawing explanation

Fig. 1 is the inventive method diagram of circuit;

Fig. 2 is that great-jump-forward reenters process and on-line parameter discrimination method schematic diagram;

Fig. 3 is the inventive method and existing methodical 8000km voyage lunar exploration airship impact accuracy comparison diagram; Fig. 3 (a) is the inventive method parameter identification figure; Fig. 3 (b) is existing method parameter identification figure.

Detailed description of the invention

Assuming that certain lunar exploration airship obtains lift by barycenter bias configuration mode reentering in process, 1ift-drag ratio is 0.3 ~ 0.4, and the controlling quantity of guidance system is angle of heel ν.The process that reenters is from height 120km place, and now airship speed is 11km/s, and local speed inclination angle is-5.8 °.

Airship adopts inertial navigation system, can measure the apparent acceleration obtained in airship body system of axes with the attitude angular velocity of airship relative to inertial coordinates system resolve through navigationsystem, airship can be obtained and returning the position in system of axes, speed, obtain the information such as the geographic latitude of airship, longitude, geodetic altitude, the angle of attack, angle of side slip, speed inclination angle simultaneously.

As shown in Figure 2, for the airship of the present embodiment, specific embodiment of the invention method is as follows:

Airship highly drops in 85km high process by 120km, does not implement to control, make angle of heel ν=180 °, be beneficial to air and catch airship.When airship height drops to 85km, start reentry guidance, note H _max=85km.Guidance process performs in the steps below:

Step S1. order once reenters the initial estimate of descent stage atmospheric density factor of proportionality 1ift-drag ratio initial estimate ${\left(\frac{L}{D}\right)}_{est}^{\&lsqb;0,1\&rsqb;}={\left(\frac{L}{D}\right)}_0^{\&lsqb;0,1\&rsqb;},{\left(\frac{L}{D}\right)}_0^{\&lsqb;0,1\&rsqb;}$ For the standard 1ift-drag ratio of current time.

Step S2. is by first guidance period start time t of reentry guidance _1,1to once reentering a t that jumps up _{n, 1}, at any one guidance period start time t of centre _{k, 1}in, perform following steps:

Step S2.1 utilizes the accelerometer measures that airship is installed to obtain apparent acceleration will project to the velocity vector of airship relative to the earth direction, obtain the observed reading D of drag acceleration _{m_k, 1}, computing formula is

$D_{m\_k,1}=-\frac{{\stackrel{\&RightArrow;}{v}}_{k,1}}{v_{k,1}}\&CenterDot;{\stackrel{\&RightArrow;}{W}}_{m\_k,1}---\left(1\right)$

In above formula, v _{k, 1}for t _{k, 1}the velocity magnitude of moment aircraft.Formula (2) is utilized to calculate the observed reading L of lift acceleration/accel _{m_k, 1}

$L_{m\_k,1}=\sqrt{{\stackrel{\&RightArrow;}{W}}_{m\_k,1}\&CenterDot;{\stackrel{\&RightArrow;}{W}}_{m\_k,1}-D_{m\_k,1}^2}---\left(2\right)$

Obtain the observed reading of 1ift-drag ratio thus

The computing formula of step S2.2 drag acceleration is

$D=\frac{C_D{\mathrm{\&rho;v}}^{2}S}{2m}---\left(3\right)$

Wherein, v is the speed of airship relative to the earth, and m is airship quality, and S is airship reference area, and ρ is atmospheric density, C _dfor drag coefficient.According to the observed reading D of formula (3) and drag acceleration _{m_k, 1}, calculate the estimated valve ρ of atmospheric density _{m_k, 1}

${\mathrm{\&rho;}}_{m\_k,1}=\frac{2{\mathrm{mD}}_{m\_k,1}}{C_{D0\_k,1}{\mathrm{Sv}}_{k,1}^2}----\left(4\right)$

C _{d0_k, 1}for first guidance period start time t of reentry guidance _1,1to once reentering a t that jumps up _{n, 1}, at any one guidance period start time t of centre _{k, 1}time airship pneumatic drag coefficient.Adopt US1976 ARDC model atmosphere ARDC, according to the height H of current time _{k, 1}, can in the hope of nominal atmospheric density ρ _{0_k, 1}, and then obtain the atmospheric density factor of proportionality of current time

$K_{\mathrm{\&rho;}m\_k,1}=\frac{{\mathrm{\&rho;}}_{m\_k,1}}{{\mathrm{\&rho;}}_{0\_k,1}}---5$

Step S2.3 utilizes low-pass first order filter, obtains the estimated valve of atmospheric density factor of proportionality

$K_{\mathrm{\&rho;}\_est}^{\&lsqb;k,1\&rsqb;}=(1-K_{g\mathrm{\&rho;}})K_{\mathrm{\&rho;}m\_k,1}+K_{g\mathrm{\&rho;}}{K}_{\mathrm{\&rho;}\_est}^{\&lsqb;k-1,1\&rsqb;}---\left(6\right)$

Utilize low-pass first order filter, obtain the estimated valve of airship 1ift-drag ratio

${\left(\frac{L}{D}\right)}_{est}^{\&lsqb;k,1\&rsqb;}=(1-K_{gLD}){\left(\frac{L}{D}\right)}_{m\_k,1}+K_{gLD}{\left(\frac{L}{D}\right)}_{est}^{\&lsqb;k-1,1\&rsqb;}---\left(7\right)$

K _{g ρ}and K _gLDfor filtering gain, be taken as 0.4.Ask the estimated valve of 1ift-drag ratio factor of proportionality

$K_{LD\_est}^{\&lsqb;k,1\&rsqb;}={\left(\frac{L}{D}\right)}_{est}^{\&lsqb;k,1\&rsqb;}/{\left(\frac{L}{D}\right)}_0^{\&lsqb;k,1\&rsqb;}---\left(8\right)$

Step S2.4 utilizes numerical prediction-correction method of guidance, the flight path of forecast airship and offset landings, and is guidanceed command by feedback offset landings.In flight path forecasting process, think that atmospheric density factor of proportionality and 1ift-drag ratio factor of proportionality are constant, numerical value equals the identifier of current time.Thus, after the current guidance cycle, the computing formula of the n-th guidance cycle internal resistance acceleration/accel is

$D_{pre\_k+n}=K_{\mathrm{\&rho;}\_est}^{\&lsqb;k,1\&rsqb;}\frac{C_{D0\_k+n}{\mathrm{\&rho;}}_{0\_k+n}{v}_{k+n}^{2}S}{2m}---\left(9\right)$

The computing formula of lift acceleration/accel is

$L_{pre\_k+n}=K_{LD\_est}^{\&lsqb;k,1\&rsqb;}{K}_{\mathrm{\&rho;}\_est}^{\&lsqb;k,1\&rsqb;}\frac{C_{L0\_k+n}{\mathrm{\&rho;}}_{0\_k+n}{v}_{k+n}^{2}S}{2m}---\left(10\right)$

Following data are stored on boat-carrying computing machine by step S2.5: current time t _{k, 1}, present level H _{k, 1}with the density content factor of current time

Circulation performs step S2.1 ~ S2.5, until rate of change in altitude is 0, also namely arrives and jumps up a little.

When step S3. arrives jumping up of reentry trajectory, note present level is H _min.With t _{k, 1}moment is example, adopts height H _{k, 1}the five rank fitting of a polynomial density content factors also namely following relation is met between two parameter

$K_{r\_est}^{\&lsqb;k,1\&rsqb;}=a_0+a_1{H}_{1,k}+a_2{H}_{1,k}^2+a_3{H}_{1,k}^3+a_4{H}_{1,k}^4+a_5{H}_{1,k}^5---\left(11\right)$

In formula, a ₀~ a ₅for fitting coefficient.Make A=[a ₀, a ₁, a ₂, a ₃, a ₄, a ₅], and by the high degree of sequence H that stores in step S2 and density content factor sequence K _{r_est}be expressed as matrix form, namely

$K_{\mathrm{\&rho;}\_est}=\&lsqb;K_{\mathrm{\&rho;}\_est}^{\&lsqb;1,1\&rsqb;},K_{\mathrm{\&rho;}\_est}^{\&lsqb;2,1\&rsqb;},...,K_{\mathrm{\&rho;}\_est}^{\&lsqb;k,1\&rsqb;},...,K_{\mathrm{\&rho;}\_est}^{\&lsqb;N,1\&rsqb;}\&rsqb;---\left(12\right)$

$H=\left[\begin{array}{cccc}1& 1& \mathrm{...}& 1\\ H_{1,1}& H_{2,1}& & H_{N,1}\\ H_{1,1}^2& H_{2,1}^2& & H_{N,1}^2\\ H_{1,1}^3& H_{2,1}^3& \mathrm{...}& H_{N,1}^3\\ H_{1,1}^4& H_{2,1}^4& & H_{N,1}^4\\ H_{1,1}^5& H_{2,1}^5& \mathrm{...}& H_{N,1}^5\end{array}\right]---\left(13\right)$

Thus, formula (11) can be expressed as matrix equation

K _{ρ _ est}=AH (14) is because the data stored in step S2 are generally far more than 6 groups, therefore above formula is overdetermined equation, tries to achieve fitting coefficient A by method of least square

A＝K _{ρ_est}·H ^T(HH ^T) ^-1

(15)

Step S4. resets guidance algorithm, makes the initial estimate of airship 1ift-drag ratio ${\left(\frac{L}{D}\right)}_{est}^{\&lsqb;0,2\&rsqb;}={\left(\frac{L}{D}\right)}_{est}^{\&lsqb;N,1\&rsqb;}.$

Step S5. once reenters the ascent stage, by first guidance cycle t _1,2h is arrived to flying height _maxsome t _{n, 2}time range in, centre any one guidance cycle t _{k, 2}in, perform following steps:

Step S5.1 utilizes the accelerometer measures that airship is installed to obtain apparent acceleration will project to the velocity vector of airship relative to the earth direction, utilize formula (1) to obtain the observed reading D of drag acceleration _{m_k, 2}, utilize formula (2) to obtain the observed reading L of lift acceleration/accel _{m_k, 2}, obtain the observed reading of 1ift-drag ratio thus according to formula (7), obtain the estimated valve of airship 1ift-drag ratio according to formula (8), try to achieve 1ift-drag ratio factor of proportionality $K_{LD\_est}^{\&lsqb;k,2\&rsqb;}.$

Step S5.2 utilizes numerical prediction-correction method of guidance, the flight path of forecast airship and offset landings, and is guidanceed command by feedback offset landings.In forecasting process, think identical height H _kthe atmospheric density factor of proportionality of place's correspondence is identical, be then in [H to altitude range _min, H _max] in height value H _k, the atmospheric density factor of proportionality of its correspondence is

$K_{\mathrm{\&rho;}\_est}^{\&lsqb;k\&rsqb;}=a_0+a_1{H}_k+a_2{H}_k^2+a_3{H}_k^3+a_4{H}_k^4+a_5{H}_k^5---\left(16\right)$

Work as H _k>H _maxtime, ignore atmospheric action, think that airship is in extraatmospheric Kepler's section.

Work as H _k<H _mintime, think atmospheric density factor of proportionality and height H _minthat locates is identical, namely

$K_{\mathrm{\&rho;}\_est}^{\&lsqb;k\&rsqb;}=K_{\mathrm{\&rho;}\_est}^{\&lsqb;H=H_{\min }\&rsqb;}---\left(17\right)$

Then height H after the current guidance cycle _kthe computing formula of place's drag acceleration is

$D_{pre\_H_k}=K_{\mathrm{\&rho;}\_est}^{\&lsqb;H=H_k\&rsqb;}\frac{C_{D0\_H_k}{\mathrm{\&rho;}}_{0\_H_k}{v}_{H_k}^{2}S}{2m}---\left(18\right)$

The computing formula of lift acceleration/accel is

$L_{pre\_H_k}=K_{LD\_est}^{\&lsqb;k,2\&rsqb;}{K}_{\mathrm{\&rho;}\_est}^{\&lsqb;H=H_k\&rsqb;}\frac{C_{L0\_H_k}{\mathrm{\&rho;}}_{0\_H_k}{v}_{H_k}^{2}S}{2m}---\left(19\right)$

Step S6. works as airship and jumps up height higher than H _maxafter=85km, think that airship has flown out atmospheric envelope, enter Kepler's section, guidance system quits work.Arrive the vertex of segmental arc of jumping up when airship and again drop to H _maxafter=85km, start secondary reentry guidance section.

At the secondary reentry guidance section initial stage, (airship height is greater than H to step S7. _min), the algorithm of guidance system is identical with step S5: i.e., in the track forecasting process of numerical prediction-correction guidance, think at H _min≤ H≤H _maxscope in, atmospheric density scale factor K _ρwith the relation of height H with once to reenter descent stage identical, (16) formula is adopted to obtain the estimated valve of atmospheric density factor of proportionality; Work as H<H _mintime, think atmospheric density factor of proportionality and H _minlocate identical.Airship 1ift-drag ratio factor of proportionality is assumed to constant value, numerically equals the estimated valve of current time.

In the secondary reentry guidance section later stage, (airship height is less than H to step S8. _min), get the height H of airship parachute-opening point _end=10km, the algorithm of guidance system is identical with step S2: i.e. in the track forecasting process of numerical prediction-correction guidance, think that the value of atmospheric density factor of proportionality and airship 1ift-drag ratio factor of proportionality is constant value in this height section, numerically equal the identifier of current time, therefore according to atmospheric density factor of proportionality and the 1ift-drag ratio factor of proportionality of formula (6) and (8) real-time identification current time.

For further illustrating the effect of the present invention to lunar exploration airship great-jump-forward reentry guidance algorithm, provide a simulation example herein.Simulated conditions is set to: the height that airship arrives reentry point is 120km, and reentry velocity is 11km/s, and reentry angle is-5.8 °, and aerodynamic parameter is with reference to U.S. member explorer vehicle (CEV), and standard voyage is 8000km.

In simulation process, consider lift coefficient, drag coefficient and atmospheric density deviation.Table 1 gives and does not adopt on-line parameter identification algorithm of the present invention, also namely uses merely Largest Single Item bias contribution during numerical prediction-correction method of guidance.

The Largest Single Item bias contribution of the simple numerical prediction-correction guidance algorithm of table 1

Analytical table 1 finds: except the situation of lift coefficient bigger than normal 20% can direct into except near object point, the offset landings of all the other situations all cannot meet mission requirements.Especially the situation of atmospheric density on the low side 35%, airship can not be caught by earth atmosphere, reenters mission failure.

Assuming that lift coefficient, drag coefficient, atmospheric density error all obey zero-mean normal distribution, relative error mean square error is respectively 20%, 20%, 35%.Fig. 3, based on MonteCarlo numerical value target practice analog simulation method, gives and whether adopts guidance precision Comparative result of the present invention.As we can see from the figure: adopt after the present invention, the air all realizing airship is caught, and vertical journey deviation basic controlling within 10km.Therefore, the present invention can effectively increase great-jump-forward reentry guidance convergence and precision.

Claims (1)

1. the on-line parameter discrimination method that reenters of lunar exploration airship great-jump-forward, it is characterized in that, the method is:

1) when lunar exploration airship height drops to H _maxtime, start reentry guidance, the initial estimate of setting atmospheric density factor of proportionality with the initial estimate of airship 1ift-drag ratio

2) from first guidance period start time t of reentry guidance _1,1to once reentering a moment t that jumps up _{n, 1}time period in, any one guidance period start time t in this time period _{k, 1}, perform following steps:

2a) set current time as t _{k, 1}, measure and obtain t _{k, 1}the apparent acceleration of moment lunar exploration airship lunar exploration airship t is obtained according to inertial navigation principle _{k, 1}the flying height H in moment _{k, 1}and rate of change in altitude according to calculate t _{k, 1}the observed reading D of moment lunar exploration airship drag acceleration _{m_k, 1}with the observed reading L of lift acceleration/accel _{m_k, 1}, and then obtain t _{k, 1}the observed reading of moment 1ift-drag ratio

2b) by the observed reading D of drag acceleration _{m_k, 1}calculate t _{k, 1}the estimated valve ρ of moment atmospheric density _{m_k, 1}, according to lunar exploration airship t _{k, 1}the level height H in moment _{k, 1}obtain t _{k, 1}the atmospheric density ρ of moment nominal Atmospheric models _{0_k, 1}, thus obtain lunar exploration airship t _{k, 1}the atmospheric density factor of proportionality in moment

The high frequency noise of 2c) filtering atmospheric density factor of proportionality and airship 1ift-drag ratio, obtains t _{k, 1}the estimated valve of moment atmospheric density factor of proportionality with the estimated valve of lunar exploration airship 1ift-drag ratio

$K_{\mathrm{\&rho;}\_\mathrm{est}}^{[k,1]}=(1-K_{\mathrm{g\&rho;}})K_{\mathrm{\&rho;m}\_k,1}+K_{\mathrm{g\&rho;}}{K}_{\mathrm{\&rho;}\_\mathrm{est}}^{[k-\mathrm{1,1}]};$

${\left(\frac{L}{D}\right)}_{\mathrm{est}}^{[k,1]}=(1-K_{\mathrm{gLD}}){\left(\frac{L}{D}\right)}_{m\_k,1}+K_{\mathrm{gLD}}{\left(\frac{L}{D}\right)}_{\mathrm{est}}^{[k-\mathrm{1,1}]};$

In above formula, for t _{k, 1}a upper moment t in moment _k-1,1the estimated valve of moment atmospheric density factor of proportionality, initial value be k _{g ρ}for atmospheric density filtering factor; for t _k-1,1the estimated valve of moment lunar exploration airship 1ift-drag ratio, initial value be k _gLDfor 1ift-drag ratio filtering factor;

Then t _{k, 1}the 1ift-drag ratio factor of proportionality in moment computing formula be:

$K_{\mathrm{LD}\_\mathrm{est}}^{[k,1]}={\left(\frac{L}{D}\right)}_{\mathrm{est}}^{[k,1]}/{\left(\frac{L}{D}\right)}_0^{[k,1]}$

In above formula, for t _{k, 1}the nominal 1ift-drag ratio of the lunar exploration airship that the moment is corresponding, can be obtained by the Aerodynamic Coefficient that lunar exploration airship stores;

2d) store t _{k, 1}the flying height H of moment lunar exploration airship _{k, 1}with the estimated valve of atmospheric density factor of proportionality

2e) utilize numerical prediction-correction method of guidance to guide, forecasting process supposition atmospheric density factor of proportionality and 1ift-drag ratio factor of proportionality are constant value, obtain reentering lift acceleration/accel L in process _prewith drag acceleration D _preestimated valve; By dynam integration, forecast the flight path of lunar exploration airship, and then revise offset landings; In a forecasting process, from current time t _{k, 1}to the drag acceleration D reentering lunar exploration airship in end time _{pre_k+n}computing formula be:

$D_{\mathrm{pre}\_k+n}=K_{\mathrm{\&rho;}\_\mathrm{est}}^{[k,1]}\frac{C_{D0\_k+n}{\mathrm{\&rho;}}_{0\_k+n}{v}_{k+n}^{2}S}{2m};$

In above formula, m is the quality of lunar exploration airship; S is lunar exploration airship reference area; Subscript k+n represents from current time t _{k, 1}the n-th guidance cycle after rising, ρ _{0_k+n}for the atmospheric density obtained by ARDC model atmosphere ARDC, C _{d0_k+n}for the pneumatic drag coefficient that the aerodynamic data being flown loading vessel by lunar exploration obtains, v _k+nfor the speed of lunar exploration airship; In a forecasting process, from current t _{k, 1}moment is to the lift acceleration/accel L reentering lunar exploration airship in end time _{pre_k+n}computing formula be:

$L_{\mathrm{pre}\_k+n}=K_{\mathrm{LD}\_\mathrm{est}}^{[k,1]}{K}_{\mathrm{\&rho;}\_\mathrm{est}}^{[k,1]}\frac{C_{L0\_k+n}{\mathrm{\&rho;}}_{0\_k+n}{v}_{k+n}^{2}S}{2m};$

Wherein, C _{l0_k+n}for the aerodynamic lift coefficient obtained by the aerodynamic data that lunar exploration airship stores;

2f) repeat above-mentioned steps 2a) ~ 2e), until the rate of change in altitude of lunar exploration airship be 0, namely lunar exploration airship arrive reentry trajectory jump up a little, enter step 3);

3) lunar exploration airship is recorded in described height H of jumping up a little _min, according to above-mentioned steps 2e) in from first of reentry guidance guidance period start time t _1,1to once reentering a moment t that jumps up _{n, 1}time period in store lunar exploration airship altitude information H and atmospheric density factor of proportionality estimated valve data K _{ρ _ est}, utilize the estimated valve K of following formula matching height H and atmospheric density factor of proportionality _{ρ _ est}between relation:

K _{ρ_est}=A·H；

Wherein

$H=\left[\begin{array}{cccc}1& 1& ...& 1\\ H_{\mathrm{1,1}}& H_{\mathrm{2,1}}& & H_{N,1}\\ H_{\mathrm{1,1}}^2& H_{\mathrm{2,1}}^2& & H_{N,1}^2\\ H_{\mathrm{1,1}}^3& H_{\mathrm{2,1}}^3& ...& H_{N,1}^3\\ H_{\mathrm{1,1}}^4& H_{\mathrm{2,1}}^4& & H_{N,1}^4\\ H_{\mathrm{1,1}}^5& H_{\mathrm{2,1}}^5& ...& H_{N,1}^5\end{array}\right];$

$K_{\mathrm{\&rho;}\_\mathrm{est}}=[K_{\mathrm{\&rho;}\_\mathrm{est}}^{\left[\mathrm{1,1}\right]},K_{\mathrm{\&rho;}\_\mathrm{est}}^{\left[\mathrm{2,1}\right]},...,K_{\mathrm{\&rho;}\_\mathrm{est}}^{[N,1]}];$

A=[a ₀, a ₁, a ₂, a ₃, a ₄, a ₅], a ₀~ a ₅for fitting coefficient, can solve according to method of least square

A=K _{ρ_est}·H ^T(HH ^T) ^-1

4) setting the initial estimate once reentering ascent stage lunar exploration airship 1ift-drag ratio is

5) once the ascent stage is reentered at lunar exploration airship, by first guidance period start time t _1,2h is arrived to lunar exploration airship flying height _maxmoment t _{n, 2}time period in, any instant t in this time period _{k, 2}in, perform following steps:

The apparent acceleration of lunar exploration airship 5a) is obtained by inertial navigation system measurement according to calculate the observed reading D of lunar exploration airship drag acceleration _{m_k, 2}with the observed reading L of lift acceleration/accel _{m_k, 2}, and then obtain the observed reading of 1ift-drag ratio the high frequency noise of filtering 1ift-drag ratio observed reading, obtains the estimated valve of lunar exploration airship 1ift-drag ratio

${\left(\frac{L}{D}\right)}_{\mathrm{est}}^{[k,2]}=(1-K_{\mathrm{gLD}}){\left(\frac{L}{D}\right)}_{m\_k,2}+K_{\mathrm{gLD}}{\left(\frac{L}{D}\right)}_{\mathrm{est}}^{[k-1,2]}$

In above formula, for t _{k, 2}a upper moment t in moment _k-1,2moment lunar exploration airship 1ift-drag ratio estimated valve, initial value be k _gLDfor 1ift-drag ratio filtering factor, for t _{k, 2}moment lunar exploration airship 1ift-drag ratio observed reading; Then t _{k, 2}moment 1ift-drag ratio factor of proportionality estimated valve for t _{k, 2}the nominal 1ift-drag ratio of the lunar exploration airship that the moment is corresponding;

5b) following formula is utilized to solve t _{k, 2}moment atmospheric density factor of proportionality estimated valve

$K_{\mathrm{\&rho;}\_\mathrm{est}}^{[k,2]}=a_0+a_1{H}_{k,2}+a_2{H}_{k,2}^2+a_3{H}_{k,2}^3+a_4{H}_{k,2}^4+a_5{H}_{k,2}^5;$

H _{k, 2}for t _{k, 2}the flying height of moment lunar exploration airship; Work as H _k>H _maxtime, ignore atmospheric influence; Work as H _{k, 2}<H _mintime, for lunar exploration airship is H once reentering a height of jumping up _minthe estimated valve of place's atmospheric density factor of proportionality;

5c) utilizing numerical prediction-correction method of guidance, assuming that 1ift-drag ratio factor of proportionality is constant value, utilizing following formula to estimate in a forecasting process from current time to reentering the drag acceleration before end

$D_{{\mathrm{pre}\_H}_k}=K_{\mathrm{\&rho;}\_\mathrm{est}}^{H_k}\frac{C_{{D0\_H}_k}{\mathrm{\&rho;}}_{{0\_H}_k}{v}_{H_k}^{2}S}{2m};$

for present level H _kunder pneumatic drag coefficient, can be obtained by aerodynamic data that airship stores; for present level H _kthe atmospheric density at place, can calculate according to ARDC model atmosphere ARDC; for lunar exploration airship speed, can be provided by navigationsystem; for present level H _kunder the estimated valve of atmospheric density factor of proportionality;

Following formula is utilized to estimate in a forecasting process from current time to reentering the lift acceleration/accel before end

$L_{{\mathrm{pre}\_H}_k}={K_{\mathrm{LD}\_\mathrm{est}}^{[k,2]}K}_{\mathrm{\&rho;}\_\mathrm{est}}^{H_k}\frac{C_{{L0\_H}_k}{\mathrm{\&rho;}}_{{0\_H}_k}{v}_{H_k}^{2}S}{2m};$

for present level H _kthe aerodynamic lift coefficient that place is corresponding, can be obtained by the aerodynamic data that airship stores;

The lunar exploration airship drag acceleration obtained will be estimated with lift acceleration/accel substitute in kinetics equation and carry out integration, obtain the lunar exploration airship flight path predicted, and guidance command reduction offset landings by what revise current time;

6) jump up when lunar exploration airship and be highly greater than H _maxtime, guidance system quits work; Arrive the vertex of the section of jumping up when lunar exploration airship and again drop to H _maxafter, secondary reentry guidance section starts, and at the secondary reentry guidance section initial stage, namely lunar exploration airship height is greater than H _mintime, guidance system is according to the works of step 5); In the secondary reentry guidance section later stage, namely lunar exploration airship height is less than H _mintime, guidance system is according to step 2) works, until lunar exploration airship arrives parachute-opening point height H _end.