CN103224023B - Phase plane self-adaptation control method based on characteristic model - Google Patents

Abstract

Disclosed is a phase plane self-adaptation control method based on a characteristic model. The method comprises the following steps: (1) a speed limit value in a jet control law is designed; (2) an angular speed maximum value of a stepped zone, an angular speed maximum value of a low thrust zone, a dead zone theta D and a stepped threshold value theta v in the jet control law are designed; (3) a high thrust zone threshold value theta B in the jet control law is designed; (4) a low thrust angular speed accelerated speed parameter ac2 and a high thrust angular speed accelerated speed parameter ac1 in the jet control law are calculated according to a golden ratio coefficient; (5) a stepped zone parameter kj in the jet control law is calculated according to the low thrust angular speed accelerated speed parameter ac2, and a parabola coefficient KX in the jet control law is calculated according to the high thrust angular speed accelerated speed parameter ac1 and other phase plane parameters; (6) the controlled quantity is calculated according to parameters designed in the above-mentioned steps and a phase plane jet control logic, namely, the jet length of an engine is confirmed, and the engine is controlled within a sampling control period according to the confirmed control quantity.

Description

A kind of phase plane self-adaptation control method of feature based model

Technical field

The present invention relates to a kind of spacecraft phase plane control method, particularly for Spacecraft, in spacecrafts rendezvous process, there is high precision, the stable gas puff Z-pinch method when interference such as plume, Time Delay of Systems.

Background technology

Technique in Rendezvous and Docking refers to that two spacecrafts are joined (intersection) by preposition, speed and time in orbit, then through attitude tracking, draw close until the thru-flight course of action of structurally link into an integrated entity (docking).Carry out two spacecrafts of Technique in Rendezvous and Docking, usual one is referred to as target aircraft (abbreviation object machine), and another is referred to as follows the trail of aircraft (abbreviation tracker).In spacecrafts rendezvous process, tracker is initiatively, the general spacecrafts rendezvous realizing two spacecrafts by changing tracker relative to the position of object machine and attitude stage by stage.Spacecrafts rendezvous process generally divides four-stage: long distance navigation section, target-seeking section, Approach phase, to draw close and Butt Section.

In spacecrafts rendezvous process, tracker needs to be started shooting by driving engine frequently to carry out track and attitude maneuver, causes propellant expenditure, thus causes tracker mass property and Inertia Characteristics constantly to change; Track and attitude maneuver also can excite solar array flexible vibration, and this windsurfing flexible vibration characteristic itself exists uncertain; The plume that driving engine start produces is beaten and can produce to tracker the exciting force and disturbance torque that change with windsurfing corner on solar array; When this external relative distance is nearer there is coupling in the control of relative attitude and relative position, and control system life period postpones etc., and this all makes the accurate control of spacecrafts rendezvous become a difficult problem.

The intelligent adaptive control method of feature based model is that the prosperous academician of Wu Hong proposes for 1992, through vicennial research, Theory and applications all achieves impressive progress, define the very strong Adaptive Control Theory of the practicality of complete set and method, mainly comprise the contents such as feature modeling Theories and methods, calculus golden cut adaptive control method.See document [1] (Wu Hongxin, Hu Jun, separate Yongchun. the Intelligent Adaptive Control of feature based model. Beijing, China Science Tech Publishing House, 2009).

Document is pointed out in [1], and so-called feature modeling is exactly require to carry out modeling according to object dynamic characteristic, environmental characteristic and controller performance.Characteristic model is made up of characteristic variable and characteristic parameter.Characteristic variable is the variable of reflection object application force and motion feature, as the variable such as position, speed, acceleration/accel of controlling quantity, output.Characteristic parameter is the key parameter of reflection application force and motion feature variable relation, comprises gain, delay time, order, Relative order, coefficient and time-varying characteristics parameter thereof.The feature of characteristic model is that model considers control overflow, verified in theory: the position for Linear Time Invariant, linear time-varying, nonlinear constant and some nonlinear and time-varying system keeps and Position Tracking Control, its characteristic model can describe with low order slow time-varying difference equation; Wherein, for open-loop stable system or open-loop unstable but Relative order be less than or equal to 2 system, its characteristic model can with second order slow time-varying difference equation describe.Therefore be different from usually as in the methods such as mode truncation the way that high order mode is cast out, it incorporates high order mode in the characteristic parameter of characteristic model for information about, and not drop-out, model accuracy is high, and simpler than prime power model.The output of characteristic model and the output of practical object remain within tolerance in dynamic process, and two outputs are equal under steady state conditions.There is a unmodeled dynamiocs between characteristic model and practical object, under the condition that the sampling period is enough little, this error is approximately the infinitesimal of the same order amount in sampling period.If be that the controller of object designs has robustness to this unmodeled dynamiocs with characteristic model, so former object is just stable under this controller action.Therefore feature modeling breaches control theory bottleneck in actual applications, carry out lower order controller design for high-order complex object and Design of intelligent controller provides foundation, especially the practical application of Adaptive Control Theory provides a new technological approaches.

Golden-section adaptive control rule golden section and minimum variance adaptive control is restrained a kind of new self-adaptation control method combined, under certain condition, the stability of closed loop system and the robust stability to unmodeled dynamiocs when can ensure that parameter does not converge on " true value ", (Yongchun is separated see document [2], Wu Hongxin. the application of golden section in self-adaptive robust controller design. automation journal, 1992,18 (02): 177-185).

But the golden-section adaptive control device of feature based model is linear controller, can not be directly used in and solves the so jet nonlinear Control problem of spacecrafts rendezvous.

Phase plane controls to be classical gas puff Z-pinch method, is widely used in the gesture stability of satellite and airship.Document [3] (Tu Shancheng edits. " Satellite Attitude Dynamics and control ". Yuhang Publishing House .2001 .) chapter 13 describe the method for designing of phase plane control law.But because phase plane control design case parameter is many, and need manually to try to gather by experience, so the high precision that will realize stable spacecrafts rendezvous controls, the design of phase plane parameter is a difficult problem.

Summary of the invention

Technology of the present invention is dealt with problems and is: overcoming existing phase plane control design case parameter needs people to control examination to gather, the Parameters design that a kind of phase plane controls is provided, solves the Controller gain variations problem that in windsurfing in spacecrafts rendezvous process flexible large, plume serious interference, attitude and orbit coupling, the large situation of system delay, robustness is good, control accuracy is high, adaptable.

Technical solution of the present invention is: a kind of phase plane self-adaptation control method of feature based model, and step is as follows:

(1) according to system delay Δ T _delaywith the speed limit in dynamic performance requirements design gas puff Z-pinch rule

(2) according to dynamic performance requirement, and the speed limit that step (1) designs is considered stepping district cireular frequency maxim in design gas puff Z-pinch rule with low thrust district cireular frequency maxim and according to measured error θ _errorwith system delay Δ T _delaydead band threshold values θ in design gas puff Z-pinch rule _dwith stepping threshold values θ _v;

(3) according to the dead band threshold values θ determined in control accuracy requirement and step (2) _dhigh thrust district threshold values θ in design gas puff Z-pinch rule _b;

(4) according to low thrust angular acceleration parameter a in golden section coefficient calculations gas puff Z-pinch rule _c2and high thrust angular acceleration parameter a _c1;

(5) according to low thrust angular acceleration parameter a _c2calculate the stepping district parameter k in gas puff Z-pinch rule _j; According to high thrust angular acceleration parameter a _c1and the parabolic coefficient K in other phase plane calculation of parameter gas puff Z-pinch rule _x;

(6) according to the parameter designed in above-mentioned five steps, according to phase plane gas puff Z-pinch logical calculated controlling quantity, namely determine the jet length of driving engine, according to determined controlling quantity, driving engine is controlled within this controlling of sampling cycle.

Low thrust angular acceleration parameter in described step (4)

Wherein, k ₂∈ [0.5,0.8], k ₁∈ [0.2,0.5], Δ T is the controlling of sampling cycle.

Described k ₂optimum is k ₁optimum is

Stepping district parameter k in described step (5) _j=(1-k ₂) a _c2Δ T/ (θ _b-θ _e);

Wherein, Δ T is the controlling of sampling cycle, θ _efor being slightly less than θ _dconstant.

Parabolic coefficient K in described step (5) _x=γ ^a _jL/ a _c1, γ span [1,6], a _jLfor actual high thrust angular acceleration.

In the spacecrafts rendezvous section of drawing close, dead band threshold values θ _dwith high thrust district threshold values θ _bautomatically adjust according to following rule:

${\mathrm{\θ}}_D={\mathrm{\θ}}_{\mathrm{Dfianal}}+({\mathrm{\θ}}_{\mathrm{Dinitial}}-{\mathrm{\θ}}_{\mathrm{Dfianal}})\frac{\left(\right|X|-X_{\mathrm{fianal}})}{(X_{\mathrm{initial}}-X_{\mathrm{fianal}})}$

${\mathrm{\θ}}_B={\mathrm{\θ}}_{\mathrm{Bfinal}}+({\mathrm{\θ}}_{\mathrm{Binitial}}-{\mathrm{\θ}}_{\mathrm{Bfinal}})\frac{\left(\right|X|-X_{\mathrm{fianal}})}{(X_{\mathrm{initial}}-X_{\mathrm{fianl}})}$

Wherein, θ _dinitialfor the section of drawing close initial position dead band threshold values, θ _dfinalfor terminal location dead band threshold values, θ _binitialfor initial position high thrust district threshold values, θ _bfinalfor terminal location high thrust district threshold values, X _initialfor the absolute value of the longitudinal relative distance of initial position, X _finalfor the absolute value of the longitudinal relative distance of terminal location, X represents longitudinal relative distance.

The present invention compared with prior art beneficial effect is:

(1) spacecraft with flexible windsurfing belongs to higher-order plant.For this controlled object, when there is strong jamming and system delay, if the design of phase plane controling parameters is improper, be easy to excite windsurfing flexible vibration, cause control system unstable, the very stubborn problem faced when this problem is the Control System Design of airship spacecrafts rendezvous.This problem, through for many years to the comparative studies of distinct methods, is just solved well until the design philosophy that golden section controls is incorporated into after phase plane controls by we.Its theoretical foundation is: document [2] has demonstrated the robust stability of linear golden-section adaptive control device, and golden section controls to belong to Cautious control, is not easy to excite windsurfing flexible vibration, therefore effectively can solves windsurfing flexible vibration problem.

(2) theoretical according to the golden-section adaptive control of feature based model, carry out project navigator to the controling parameters in phase plane gas puff Z-pinch rule, controling parameters each other relation is clear and definite, and in prior art, each parameter is all tried to gather by experience one by one.

(3) consider system delay in the design of controling parameters, therefore can effectively solve system delay problem.

(4) be conducive to according to the phase plane parameter in characteristic parameter adaptive adjustment gas puff Z-pinch rule the requirement that control system meets different control accuracy in motion process.

We are according to the golden-section adaptive control theory calculate phase plane parameter of feature based model, revise, propose the phase plane self-adaptation control method of feature based model according to relative distance to controling parameters.The method has the advantages such as control accuracy is high, consumption of fuel is little, robust comformability is good, antijamming capability is strong, in the Technique in Rendezvous and Docking of No. eight, divine boat, No. nine, divine boat and No. one, Heavenly Palace close to the section of drawing close successful Application.

Accompanying drawing explanation

Fig. 1 is the phase plane self-adaptation control method design flow diagram that the present invention is based on characteristic model;

Fig. 2 is the phase plane control partition figure that the phase plane self-adaptation control method that the present invention is based on characteristic model uses;

Fig. 3 is spacecrafts rendezvous closed loop control system functional block diagram.

Detailed description of the invention

Below in conjunction with accompanying drawing, the present invention will be further described.

The phase plane self-adaptation control method of feature based model of the present invention, key content is the method for designing of phase plane parameter, is also included in some performance figure and requires, in the task of change, according to characteristic parameter, phase plane parameter to be carried out to the method for accommodation.Theoretical foundation of the present invention is that the golden-section adaptive control of feature based model is theoretical.

First the system that this method is suitable for once simply is introduced, adopt the tracker spacecrafts rendezvous closed loop control system functional block diagram of gas puff Z-pinch as shown in Figure 3.In Fig. 3, each several part implication is as follows:

Track, attitude dynamics had both comprised tracker absolute orbit and attitude dynamics, and comprising again tracker relative to the relative position of object machine and attitude dynamics, is the controlled object in control system.Dynamic disturbance is except comprising the ambient interference such as orbit perturbation, atmospheric drag force and moment, and the plume produced when also comprising driving engine start disturbs force and moment.Due to the uncertainty that driving engine start and windsurfing rotate, plume is caused to disturb force and moment also to have uncertainty.

The kinematic parameter of controlled object is measured by the measurement sensor be arranged on tracker, measure sensor and both comprised infrared earth sensor, star sensor, inertia measurement sensor, also comprise the spacecrafts rendezvous relative measurement sensors such as CCD optical imagery sensor, laser radar, microwave radar, satellite navigational equipment.According to measuring the output of sensor, estimated attitude, the attitude angular velocity of tracker by position and attitude estimator, and tracker is relative to status informations such as the relative position of object machine, relative velocity, relative attitude angle and relative attitude cireular frequencys.

Different Guidance Laws is selected, as multiple pulse guidance, CW guidance, Line of Sight Guidance etc. in the different stage of spacecrafts rendezvous.Require and Attitude estimation result according to the targeted attitude that Guidance Law exports, calculate the torque command needed for the deviation revising tracker actual attitude relative target attitude by attitude controller.In the translation section of drawing close, require and location estimation result according to target location, calculate the power instruction needed for the deviation revising tracker actual position relative target position by positioner; In other stage, Guidance Law directly provides the size and Orientation becoming rail velocity increment.Power, torque command and velocity increment require to determine to select which driving engine start and start duration (or jet length) by thruster command assignment algorithm.

The control actuating unit of tracker is made up of tens to tens of thrusters.Separate unit rail control thruster, because thrust is not by barycenter, can produce attitude disturbance moment, and separate unit appearance control thruster also can produce track exciting force, and therefore, gesture stability and orbits controlling are couplings.In order to accurately realize thrust on certain direction or moment, need one group of thruster to work to offset the effect of thruster to other direction simultaneously.Given control thrust and torque requirement, select the thruster combination that can realize above-mentioned requirements, and calculate the work-hours of every platform thruster in this combination, be just called command assignment in all thrusters.Concrete thruster command assignment algorithm has special document introduction.The gesture stability passage high, dynamic range is large is required for control accuracy, usually by the large and small two kinds of thrusters of configuration.

Actual control system, because measure sensor response, Signal transmissions, data processing, guidance, navigation and vehicle controL rule calculate etc. all needs the time, therefore output to thruster from measuring sensor to produce the time delay controlling force and moment generally also larger under current engineering factor.

Secondly, incorporated by reference document [3] controls simply to introduce to phase plane.Phase plane shown in Fig. 2 is divided into 14 control areas by some switching lines, and phase plane switching line, about initial point O Central Symmetry, is therefore only illustrated RHP below.The meaning of primary symbols see table 1, the switching line GC of RHP ₁c ₂the equation of D can be designed to also can replace with straight line is approximate.R1 is high thrust standard-sized sheet district, and R2 is low thrust standard-sized sheet district, and R3 is low thrust stepping district, R4 and R6 is high thrust speed limit district, and R5 is anti-outer skating area, and R7 is complete shut-down district.Phase plane R1, R2, R3, R4, R5, R6 district are deceleration area, reverse driving engine start.Correspondingly R ' 1, R ' 2, R ' 3, R ' 4, R ' 5, R ' 6 district are accelerating region, and forward driving engine is started shooting.R ' 7 is also complete shut-down district.It should be noted that, if only configure a kind of thruster, this thruster standard-sized sheet can be regarded as high thrust, and this thruster is opened N% regard low thrust as.N determines according to actual needs.

Table 1 primary symbols meaning

Embodiment 1

Performing step of the present invention is as follows:

(1) according to system delay Δ T _delaywith the speed limit in dynamic performance requirements design gas puff Z-pinch rule if system delay is Δ T _delay, high thrust angular acceleration a _jL, controller performance requires that the angular velocity range allowed is then can design speed limit in gas puff Z-pinch rule make it meet and ${\stackrel{\·}{\mathrm{\θ}}}_L=\left[{\stackrel{\·}{\mathrm{\θ}}}_{\min }{,\stackrel{\·}{\mathrm{\θ}}}_{\max }\right].$

(2) according to dynamic performance requirement, and the speed limit that step (1) designs is considered stepping district cireular frequency maxim in design gas puff Z-pinch rule with low thrust district cireular frequency maxim and according to measured error θ _errorwith system delay Δ T _delaydead band threshold values θ in design gas puff Z-pinch rule _dwith stepping threshold values θ _v;

Stepping district cireular frequency maxim with low thrust district cireular frequency maxim determine the height in R2 and R3 district, its principle of design be under the prerequisite meeting system response time requirement, raise R2 and R3 district as far as possible height to avoid exciting flexible vibration.On this basis, can design or after making phase trajectories enter RHP from Left half-plane, directly enter R3 district from R7 district, or enter R3 district again after entering R2 district from R7 district.

If measured error is θ _error, then dead band threshold values θ in gas puff Z-pinch rule can be designed _dmake it meet ${\mathrm{\θ}}_D>{\mathrm{\θ}}_{\mathrm{error}}+\mathrm{\Δ}{T}_{\mathrm{delay}}{\stackrel{\·}{\mathrm{\θ}}}_V,$ θ _v＞θ _error。

(3) according to the dead band threshold values θ determined in control accuracy requirement and step (2) _dhigh thrust district threshold values θ in design gas puff Z-pinch rule _b;

If control accuracy requires as θ _desired, then gas puff Z-pinch Lv Zhong high thrust district threshold values θ can be designed _bit is made to meet θ _d< θ _b< θ _desiredprerequisite under as far as possible large.

(4) according to low thrust angular acceleration parameter a in golden section coefficient calculations gas puff Z-pinch rule _c2and high thrust angular acceleration parameter a _c1;

Low thrust angular acceleration parameter high thrust angular acceleration parameter

Wherein, k ₂∈ [0.5,0.8], k ₁∈ [0.2,0.5], Δ T is the controlling of sampling cycle, k ₂optimum is k ₁optimum is $\frac{3-\sqrt{5}}{2}.$

(5) according to low thrust angular acceleration parameter a _c2calculate the stepping district parameter k in gas puff Z-pinch rule _j; According to high thrust angular acceleration parameter a _c1and the parabolic coefficient K in other phase plane calculation of parameter gas puff Z-pinch rule _x; k _j=(1-k ₂) a _c2Δ T/ (θ _b-θ _e), K _x=γ ^a _jL/ a _c1, γ span [1,6], θ _efor being slightly less than θ _dconstant.

(6) according to the parameter designed in above-mentioned five steps, according to phase plane gas puff Z-pinch logical calculated controlling quantity, namely determine the jet length of driving engine, according to determined controlling quantity, driving engine is controlled within this controlling of sampling cycle.

Below for deceleration area, provide the method for calculating of the jet length of calculative driving engine of each controlling of sampling cycle.

R1 district: oppositely high thrust motor standard-sized sheet, jet length is slightly larger than the controlling of sampling cycle.

R2 district: oppositely thrustor standard-sized sheet, jet length is slightly larger than the controlling of sampling cycle.

R3 district: the oppositely jet length T of thrustor _ncomputing formula:

$T_{N1}=\frac{\left|\stackrel{\widehat{\&CenterDot;}}{\mathrm{\&theta;}}\right|}{a_{c2}},$ $T_{N2}=K_j\frac{\left|\widehat{\mathrm{\&theta;}}\right|-{\mathrm{\&theta;}}_e}{a_{c2}},$ T _N＝T _N1+T _N2。

R4 district: the oppositely jet length T of high thrust motor _ncomputing formula:

R5 district: the oppositely jet length T of thrustor _ncomputing formula: T _n=T _min.

R6 district: the oppositely jet length T of high thrust motor _ncomputing formula:

R7 district: shutdown.

Embodiment 2

Provide the method for according to characteristic parameter, phase plane parameter being carried out to accommodation below.

If change the transverse axis of phase plane and the longitudinal axis into relative position and relative velocity, phase plane control method also can be applicable to translation and controls.In the spacecrafts rendezvous section of drawing close, tracker is close to object machine along docking corridor.Along with the minimizing of longitudinal relative distance, require that horizontal position control accuracy improves gradually.At this moment longitudinal relative distance can be selected to be characteristic parameter, according to this characteristic parameter, self-adaptative adjustment to be carried out, to realize control objectives to the phase plane parameter affecting control accuracy.

Phase plane parameter is designed with reference to embodiment 1.Along with the minimizing of longitudinal relative distance, horizontal position control accuracy requires to improve gradually to be mainly reflected in θ _dand θ _breduce gradually.If the section of drawing close initial position dead band threshold values θ _dbe designed to θ _dinitial, terminal location dead band threshold values θ _dbe designed to θ _dfinal, initial position high thrust district threshold values θ _bbe designed to θ _binitial, terminal location high thrust district threshold values θ _bbe designed to θ _bfinal, the absolute value of the longitudinal relative distance of initial position is X _initial, the absolute value of the longitudinal relative distance of terminal location is X _final, then moving closer in process, along with the change of longitudinal relative distance X, phase plane parameter θ _dand θ _bautomatically can adjust by following rule:

${\mathrm{\&theta;}}_D={\mathrm{\&theta;}}_{\mathrm{Dfianal}}+({\mathrm{\&theta;}}_{\mathrm{Dinitial}}-{\mathrm{\&theta;}}_{\mathrm{Dfianal}})\frac{\left(\right|X|-X_{\mathrm{fianal}})}{(X_{\mathrm{initial}}-X_{\mathrm{fianal}})}$

${\mathrm{\&theta;}}_B={\mathrm{\&theta;}}_{\mathrm{Bfinal}}+({\mathrm{\&theta;}}_{\mathrm{Binitial}}-{\mathrm{\&theta;}}_{\mathrm{Bfinal}})\frac{\left(\right|X|-X_{\mathrm{fianal}})}{(X_{\mathrm{initial}}-X_{\mathrm{fianl}})}$

The inventive method has the advantages such as control accuracy is high, consumption of fuel is little, robust comformability is good, antijamming capability is strong, in the Technique in Rendezvous and Docking of No. eight, divine boat, No. nine, divine boat and No. one, Heavenly Palace close to the section of drawing close successful Application.

In addition the present invention can be applied to a class high-order, time change, non-linear, large delay, strong jamming and the gas puff Z-pinch problem of uncertainty plant.Such as with the gesture stability of LIQUID-FILLED SATELLITE when becoming rail of flexible solar array.Also can be applicable to the gesture stability of general satellite.

The unspecified part of the present invention belongs to general knowledge as well known to those skilled in the art.

Claims (6)

1. a phase plane self-adaptation control method for feature based model, is characterized in that step is as follows:

(1) according to system delay Δ T _delaywith the speed limit in dynamic performance requirements design gas puff Z-pinch rule

(2) according to dynamic performance requirement, and the speed limit that step (1) designs is considered stepping district cireular frequency maxim in design gas puff Z-pinch rule with low thrust district cireular frequency maxim and according to measured error θ _errorwith system delay Δ T _delaydead band threshold values θ in design gas puff Z-pinch rule _dwith stepping threshold values θ _v;

(3) according to the dead band threshold values θ determined in control accuracy requirement and step (2) _dhigh thrust district threshold values θ in design gas puff Z-pinch rule _b;

(4) according to low thrust angular acceleration parameter a in golden section coefficient calculations gas puff Z-pinch rule _c2and high thrust angular acceleration parameter a _c1;

(5) according to low thrust angular acceleration parameter a _c2calculate the stepping district parameter k in gas puff Z-pinch rule _j; According to high thrust angular acceleration parameter a _c1and actual high thrust angular acceleration a _jL, parameter γ calculate gas puff Z-pinch rule in parabolic coefficient K _x; Wherein, γ span [1,6];

(6) according to the parameter designed in above-mentioned five steps, according to phase plane gas puff Z-pinch logical calculated controlling quantity, namely determine the jet length of driving engine, according to determined controlling quantity, driving engine is controlled within this controlling of sampling cycle.

2. the phase plane self-adaptation control method of a kind of feature based model according to claim 1, is characterized in that: the low thrust angular acceleration parameter in described step (4) high thrust angular acceleration parameter $a_{c1}={\stackrel{\&CenterDot;}{\mathrm{\&theta;}}}_L/\left(k_1\mathrm{\&Delta;T}\right);$

Wherein, k ₂∈ [0.5,0.8], k ₁∈ [0.2,0.5], Δ T is the controlling of sampling cycle.

3. the phase plane self-adaptation control method of a kind of feature based model according to claim 2, is characterized in that: described k ₂optimum is k ₁optimum is

4. the phase plane self-adaptation control method of a kind of feature based model according to claim 1, is characterized in that: stepping district parameter k in described step (5) _j=(1-k ₂) a _c2Δ T/ (θ _b-θ _e);

Wherein, k ₂∈ [0.5,0.8], Δ T is the controlling of sampling cycle, θ _efor being slightly less than θ _dconstant.

5. the phase plane self-adaptation control method of a kind of feature based model according to claim 1, is characterized in that: parabolic coefficient K in described step (5) _x=γ a _jL/ a _c1.

6. the phase plane self-adaptation control method of a kind of feature based model according to claim 1, is characterized in that: in the spacecrafts rendezvous section of drawing close, dead band threshold values θ _dwith high thrust district threshold values θ _bautomatically adjust according to following rule:

${\mathrm{\&theta;}}_D={\mathrm{\&theta;}}_{\mathrm{Dfinal}}+({\mathrm{\&theta;}}_{\mathrm{Dinitial}}-{\mathrm{\&theta;}}_{\mathrm{Dfinal}})\frac{\left(\right|X|-X_{\mathrm{final}})}{(X_{\mathrm{initial}}-X_{\mathrm{final}})}$

${\mathrm{\&theta;}}_B={\mathrm{\&theta;}}_{\mathrm{Bfinal}}+({\mathrm{\&theta;}}_{\mathrm{Binitial}}-{\mathrm{\&theta;}}_{\mathrm{Bfinal}})\frac{\left(\right|X|-X_{\mathrm{final}})}{(X_{\mathrm{initial}}-X_{\mathrm{final}})}$

Wherein, θ _dinitialfor the section of drawing close initial position dead band threshold values, θ _dfinalfor terminal location dead band threshold values, θ _binitialfor initial position high thrust district threshold values, θ _bfinalfor terminal location high thrust district threshold values, X _initialfor the absolute value of the longitudinal relative distance of initial position, X _finalfor the absolute value of the longitudinal relative distance of terminal location, X represents longitudinal relative distance.